Abdul El-Sayed has defeated Haley Stevens in Michigan’s Democratic Senate primary — albeit by a significantly smaller margin than preelection polling suggested. He’s beaten back $30 million in AIPAC spending and endorsements from top Democrats, including Michigan Gov. Gretchen Whitmer, who backed Stevens. Now he’s moving on to face the GOP nominee, former Rep. Mike Rogers, in a contest crucial to determining Senate control.

So the next question is: Can El-Sayed beat a Republican?

That will be a cinch, some in his camp claim. “For progressives, getting out of the primary is a far more difficult battle than winning in the general,” left-wing streamer Hasan Piker told CNN last week.

But the reality is that we’ve had very few true tests of how candidates associated with the left flank of the party — and specifically with the Bernie Sanders-backed revolt against the establishment — would fare in competitive swing-state general elections. 

Which means that this would be one of the first such tests, and that its outcome will have enormous stakes for the party’s future.

El-Sayed’s views on Israel and his prospective history-making turn as the first-ever Muslim senator have gotten much coverage. But El-Sayed is also a dyed-in-the-wool member of the Sanders faction, having been closely aligned with Sanders and Rep. Alexandria Ocasio-Cortez since his first political campaign in 2018. Both have campaigned for him this cycle.

If El-Sayed defeats Rogers convincingly (he’s personally predicting a 7-point victory as his baseline), he’ll become a megastar and even possibly a presidential contender. He’ll arguably be the first to prove that the Sanders formula — attacking billionaires, corporate influence, and the party establishment, while backing sweeping social welfare spending like Medicare-for-all — can win over voters who once backed President Donald Trump.

“If Abdul El-Sayed beats Stevens I actually think he’s very likely to be the next president,” Nathan J. Robinson, editor-in-chief of the socialist publication Current Affairs, posted on X last month. “If he wins the general, it’s over for centrist Dems.”

Conversely, if El-Sayed — who has trailed Rogers in recent polls — loses, skeptics may conclude that the left can’t appeal to the country’s true swing voters, and that nominating a leftist for president in 2028, or in downballot races, is too big of a risk. 

The core disagreement: Will the left’s approach succeed in the general election?

Over the past decade, it has become an article of faith across much of the left that their particular populist approach to economics and foreign policy would be potent in a general election.

They argue that typical Democrats have a more compromised platform and timid style, lacking the populist fire that can win over disaffected Americans. “Bernie would’ve won” in 2016, the saying goes, by using his outsider credibility and ambitious plans to appeal to the working-class voters and independents that went for Trump. Kamala Harris’s defeat, amid intense criticism on the left over her and Joe Biden’s support for Israel, further bolstered their sense that the party establishment just can’t get it done. 

Now, left factionalists are pointing to their several high-profile primary victories this year and arguing that energy and enthusiasm are on their side — and would translate to the general election too.

Many mainstream Democrats see this as naive. They believe that while talk of “socialism” and other far-left positions might work fine in deep blue cities and districts, it could well prove toxic in swing states that have voted for Trump. While public opinion on certain issues has moved in the left’s direction, especially Israel, candidates could have a tougher time on issues like crime, where El-Sayed has offered nuanced explanations of his “defund the police” rhetoric from 2020.

“Winning a Democratic primary means close to nothing when it comes to the general election, because these are functionally different voter pools,” Lakshya Jain of the center-left publication The Argument recently wrote.

Some analysts pointed to the primary results as a potential harbinger of general election struggles, with El-Sayed performing poorly against Stevens in the kinds of rural areas that have swung toward Republicans in the Trump era.

Are there past examples of Bernie-aligned candidates making it to tough general elections?

But even though the “Bernie vs. the establishment” battle has been raging on and off for the past decade, there have been very few clean tests about how Bernie-backed candidates would do in swing states.

Sanders never won the presidential nomination, and candidates aligned with him have generally preferred to run in strongly Democratic areas where, if they win the primary, the general election is a fait accompli. The now-expanding “squad” of left-wing House members that began with Reps. Ocasio-Cortez, Ilhan Omar, Rashida Tlaib, and Ayanna Pressley all fit this model.

There have, however, been a few past examples where Sanders-aligned candidates have made it to the general election in tough territory.

One is Andrew Gillum, whose 2018 bid for governor of Florida was backed by Sanders and other progressive groups, including billionaire Tom Steyer’s network. He won the nomination over establishment favorite Gwen Graham — but then went on to lose the general election to Ron DeSantis by a mere four-tenths of a percentage point, ushering in what became the era of Deep Red Florida.

In nominating Gillum, did the left blow a winnable race amid a national Democratic wave? Maybe — but one counterargument is that longtime Sen. Bill Nelson, a thoroughly mainstream Democrat, lost his seat by a similarly tight margin in that same election. Florida may have just been trending redder generally. Gillum’s post-election personal struggles also color views of his candidacy today.

That same cycle, in Nebraska, Kara Eastman upset the establishment favorite to win the nomination for a swing House district. (Sanders hadn’t endorsed her in the primary, but she ran on a platform of Medicare-for-all and free college.) She lost in 2018 — and again in 2020, even though Joe Biden won her district — and has been viewed inside the party as a cautionary tale about blowing a winnable race ever since.

The outcome was happier for Keith Ellison, a close Sanders ally who won his race to be Minnesota’s attorney general in 2018. Yet his margin of victory — about 4 percentage points — wasn’t particularly impressive in a lean blue state in a strongly Democratic year when Tim Walz won the governor’s race by 11. Later, when Ellison ran for reelection in 2022, he eked out a win by less than 1 point while Walz won by 8.

John Fetterman was generally understood to be a “Bernie faction” candidate when he ran for an open US Senate seat in Pennsylvania in 2022. (Sanders stayed neutral in the primary that year, but had backed Fetterman’s earlier lieutenant governor campaign and championed him in the general election.) But he wasn’t exactly running on a bold left platform — he backed off his past Medicare-for-all support and was pro-fracking.

Fetterman ended up winning by 5 points even though it was a favorable cycle for Republicans — but simultaneously, the more centrist Josh Shapiro won the governor’s race by 15. However, Shapiro’s opponent was believed to be weaker and more extreme, and the stroke Fetterman suffered that year complicates efforts to read too much into his performance. 

Separate from his electoral performance, Fetterman’s association with the Sanders left ended years ago as his politics moved to the right; lately, there has been talk he might switch parties. This has also fueled arguments from moderates that populist outsider candidates are less reliable than more conventional ones even if they do manage to win.

El-Sayed is running on a bold left platform in a true swing state

Taking this grab bag of outcomes together, one can see why mainstream Democrats might be skeptical that the left has the secret sauce to win swing voters. But each race has had its own unique circumstances and complications.

Furthermore, in this cycle, the electorate could be in a different place — more disenchanted by the establishment and more willing to take a chance on further-left candidates.

Graham Platner was set to be the great hope for the Sanders faction until his Senate campaign was derailed by scandal and he was replaced on the ballot by Troy Jackson — another Sanders-aligned candidate, albeit one with his own long history in state politics. But though Jackson is facing a formidable Republican incumbent, Maine is a blue state that hasn’t backed a GOP presidential candidate since 1988.

Michigan is tougher turf, having voted for Trump in two of the past three presidential elections. The conventional wisdom for how a Democrat can win a race there is to run a cautious campaign, being careful not to alienate potentially winnable Republican and independent voters. A good example is Gary Peters, the departing two-term senator El-Sayed is running to replace, who’s a traditional pro-union Democrat known for keeping a low profile in general.

El-Sayed is not doing that. He supports Medicare-for-all, a billionaire wealth tax, abolishing ICE, and a full arms embargo on Israel. While El-Sayed is not a democratic socialist, he’s very closely tied to Sanders and Ocasio-Cortez — the trio held hands and raised their arms together at a recent rally, in an image Michigan voters may well see in GOP-funded attack ads. 

And back in 2020, El-Sayed expressed sympathy with activists calling to “defund the police” and said most major cities spend far too much on police departments, though he’s tried to downplay those views this cycle.

“I’m worried that he’s more passionate about attacking other Dems than Republicans and may not be ready for what’s coming,” Adam Jentleson, president of the center-left Searchlight Institute, posted on X recently.

Now that he’s won the primary, El-Sayed will get the chance to try and show he’s ready — and to test whether his bold left platform will be his greatest asset, or a millstone.

This story appeared in The Logoff, a daily newsletter that helps you stay informed about the Trump administration without letting political news take over your life. Subscribe here.

Welcome to The Logoff: President Donald Trump’s renovation mania is reaching new heights. 

What’s happening? Not satisfied with a paved Rose Garden, lavish amounts of gold, and a granite walkway paid for with national parks money, Trump is currently adding a helipad to the White House lawn. Construction, begun in June, has been running 24/7, with the goal of finishing the project before a visit from Chinese President Xi Jinping in late September. 

But all is not well: As the Washington Post reported on Tuesday, contractors are now undoing their work as Trump demands a perfectly level helipad (on a South Lawn that is anything but).

It’s the latest evidence of something that New York Times reporter Maggie Haberman recently described in a podcast with Ezra Klein: By her estimate, around 70 percent of Trump’s “mind share” is tied up not with the Iran war, affordability, or the upcoming midterms, but “on his renovations,” which she describes as a “legacy thing.”

What else does Trump have in the works? In addition to the helipad, Trump is currently working on renovating Lafayette Square, across the street from the White House; on Monday, he mused about his “big decision” on whether to change the park’s pavers from red brick to granite.

He’s also re-gilding four statues near the Lincoln Memorial in vivid 23.75-karat gold; repairing the re-drained Reflecting Pool, which was the victim of a rush job the first time around; and moving ahead with his plan for a 250-foot arch, which could be approved by early next month and which the National Park Service recently concluded would have “adverse effects” on more than three dozen historic sites in the DC area.

What’s the big picture? Trump’s construction fervor has only accelerated since he returned to office last year, especially as other agenda items have eluded him. A midterms loss will further curtail his options for the final two years of his presidency, so don’t expect it to stop here.

And with that, it’s time to log off…

Here’s a stat I did not know, courtesy of my colleague Bryan Walsh: Traffic injuries are “the leading cause of death for everyone between the ages of 5 and 29 — ahead of malaria, ahead of war and homicide, ahead of every disease we spend more time worrying about.” 

That’s pretty grim, and not very Logoff-y. But there’s good news, too: As Bryan reports, between 2011 and 2025, global road deaths have fallen by 21 percent, even as more vehicles take to the roads — thanks in large part to a change in how we think about traffic accidents. You can read Bryan’s full story about what changed — and what we still have to work on — here with a gift link.

Have a great evening, and we’ll see you right back here tomorrow!

Welcome to The Midterms, Actually. Every week I’ll be writing about two things: one big idea or trend I see forming across the 2026 primaries, and one thing I’m hearing from a voice on the ground. Let’s get into it.

One big idea: The Uncommitted movement has already won

During the 2024 presidential election, on the third night of the Democratic National Convention in Chicago, we recorded an episode of my election podcast The Run-Up that I still think about all the time. The premise was simple: I wanted to know whether Democrats were making a material change at the top of the ticket or just a cosmetic one. So we asked convention-goers a plain question — how would Kamala Harris’s Democratic Party be any different from Joe Biden’s?

The tone of the piece reflected the optimism of the room. The activists, down to a person, thought Harris was showing a newfound openness to the administration’s critics. But underneath the hopium was a specific policy desire: that Harris would break from Biden’s unconditional support of Israel’s violent campaign in Gaza. That afternoon, I spent time with members of the Uncommitted movement, born in Michigan, who were trying to pressure the DNC into putting a single Palestinian speaker on the stage. They held press conferences with grieving families, and warned the party that the Democrats’ Muslim and Arab base in particular would not forget being sidelined. 

United Auto Workers president Shawn Fain made clear the confidence surrounding Harris at the time was tied to the belief her posture around Gaza would be different from Biden’s: “She’s made, obviously, a couple different comments on Gaza,” he said. “We called for a ceasefire in UAW back in November. And obviously, she understands the humanity issue of this. And I mean, my conversations I’ve had with Kamala Harris have been very good, very fruitful.” 

They didn’t get one. Not the speaker, not the policy shift, not the break from Biden. Harris famously refused to distance herself from her boss, whether it was Israel or anything else — and likely lost because of it. 

And while many of the most prominent voices of the Uncommitted movement (including current Michigan Senate candidate Abdul El-Sayed) ended up publicly backing Harris in the end, some of Harris’s most ardent supporters have repeatedly blamed left-wing activists for undermining her campaign. Haley Stevens, the Michigan representative who’s running for the Senate as the party’s establishment choice, has sought to make El-Sayed’s support for Uncommitted an issue in that race, suggesting that the movement planted seeds of discontent ahead of a must-win presidential election. 

I want to argue the opposite. Two years after the Uncommitted movement shook up the Democratic Party, its impact has never been more clear. Public opinion of Israel has plummeted across the American electorate, to the point where the movement’s position is now a majority opinion. Per Quinnipiac, 77 percent of Democrats say they believe Israel is committing a genocide — an unthinkable number just two years ago. The Uncommitted movement also led to the growth and maturation of the Democrats’ Muslim left, which has already molded prominent elected officials like New York City Mayor Zohran Mamdani and Michigan Senate candidate Abdul El-Sayed. 

Even more, I think Uncommitted challenged the blame-the-voters framework that allows the establishment to avoid self-reflection. The movement was never only about a DNC speaking slot, but about the idea that every vote must be earned, not owed. It was a retort to “vote blue no matter who” — an argument that conditional support for the party is not only valid, but of moral importance. 

In 2024, that posture got you branded a spoiler. Today it’s just how Democrats talk. When I sat down with Hasan Piker — the most popular voice on the Democratic left — he told me he voted for Kamala Harris but pointedly refused to endorse her, and named exactly why:

“The only reason for my refusal to openly endorse was because of…a continuation of the genocide in Gaza.”

“I understand the harm reduction arguments…but it’s not a practical approach to winning elections, and demonstrably it’s been a failure with Hillary Clinton and Kamala Harris.”

Piker put the mechanism plainly. October 7, 2023, and its aftermath, he told me, were “a seismic shift that gave a lot of confidence to the left, that as long as they stood by their positions…people would actually see the truth.” The movement didn’t move Biden. It moved the electorate Biden’s successors now have to answer to.

And you can trace that shifted ground straight into this year’s wins. In New York, Zohran Mamdani rode exactly this coalition into City Hall — a victory Piker described as the proof of concept he’d been waiting for: “that’s why I was so grateful for Zohran’s victory…I can just point to him.” In Congress, progressives like Darializa Avila Chevalier in New York City and Melat Kiros in Colorado add to the growing evidence that acknowledging Israel’s actions in Gaza as a genocide is now a litmus test for Democratic voters gauging the trust and credibility of their candidates.

Which brings us to Michigan, where the whole thing started. It’s easy to forget now, but the Uncommitted movement was born in the Michigan primary — and the general-election bill came due there, too. Kamala Harris got 36 percent of the vote in Dearborn, a city Joe Biden had carried with nearly 69 percent in 2020. She lost the state by about 80,000 votes. The voters the party told to fall in line didn’t, and Michigan is where that refusal was most legible.

Now look at who’s competitive in this week’s Senate primary. Abdul El-Sayed is running on the precise argument Uncommitted was making in the streets of Chicago — except now it’s a statewide campaign with real money behind it, against a candidate backed by tens of millions in pro-Israel spending. He turns the foreign-policy question into a pocketbook one and refuses to soften the language. As he said on America, Actually:

“There is an electoral cost to backstopping genocide and apartheid.”

“If you can’t call a genocide perpetrated by our tax dollars the moral abomination that it is, then at some point it’s hard for me to believe that you’re gonna fight for anything,” he told me.

Whether or not El-Sayed wins today, the fact that a candidate running explicitly against AIPAC and unconditional aid is fighting for a Senate seat in a swing state is the clearest possible measure of how far the ground has moved since that convention floor.

In the coming months, it may be that the Democrats’ moderate wing — think tanks like Third Way, party leadership, and those offended by the electorate’s rising leftist tide — will adopt some of the same tactics they criticized. Already, Rep. Hillary Scholten, who is supporting Stevens in Michigan, has signaled she may refuse to support El-Sayed in a general election, due to his association with Piker. 

It seems “vote blue no matter who” has its limits. For progressives. For centrists. For everyone. 

What I’m hearing

With Michigan’s primary set to conclude today, that leaves the Wisconsin governor’s race as the next big election on the Democratic calendar. This week, I reached out to Dan Shafer, a reporter in Milwaukee who has been covering the race for months. Schafer writes for the Recombobulation Area, where he’s political editor and writes a weekly opinion column. 

It looks like Francesca Hong, the DSA-aligned state assemblywoman, has a commanding lead in the Wisconsin governor’s primary. This would’ve been a shock a year ago. How did this happen?

The short answer is that Francesca Hong has simply run the best campaign. She has run circles around the consultant-brained conventional approaches of her opponents, and has found success by doing things decidedly differently.

From the very beginning, I characterized her as the “wild card” in the race. And while I certainly did not envision her having a commanding lead in the polls in the final stretch in the primary, I knew that, given her background in the restaurant and hospitality industry, she wouldn’t let anyone out-work or out-hustle her on the campaign trail. Her relentless ground game and digital organizing energy essentially issued a challenge to every other campaign to step up. The rest of the field failed that test. 

She’s also absolutely dominating this primary with younger voters. In the Marquette University Law School Poll from last week, the data showed her with the support of 66% of likely primary voters in the 18-29 age group, and 51% in the 30-44 age group. No other candidate was even in double digits. I write a breakdown of every new Marquette poll, and these are just preposterous numbers. For Gen Z and millennials, she is the candidate.

Hong has made data centers and AI backlash a centerpiece of this race. How important is this issue specifically in this race?

I think it’s been the No. 1 issue [since] she’s taken off in this primary. 

She backed a data center moratorium in January, and began running on this issue early. It’s been a very animating issue in Wisconsin, first for all the reasons the anti-data center backlash is coming to so many Great Lakes states, but also because of the failed Foxconn project in Racine County that happened during the first Trump administration under then-Gov. Scott Walker. It came with the promise of thousands and thousands of jobs, Trump called it the “eighth wonder of the world,” and it’s just been a miserable failure on so many levels. Hong just filmed her most recent ad at the Foxconn site — a project the Republican nominee Tom Tiffany voted for while in the state Senate at the time.

Politically, though, what really helped Hong on this issue is that no other candidate running backed a data center moratorium. One of the most significant moments of the primary happened in a June 3 candidate forum focused on working-class issues, where moderator JT Cestkowski asked a yes-or-no question on whether the candidates back a data center moratorium. Only Hong said yes. If any other candidate had said yes, that would have given the anti-data center voter options in the primary. Instead, they all went to Hong.

There is an assertion that Hong is a weak general election candidate. Is there evidence for this? 

This is Wisconsin. Anyone pretending a statewide election in this of all states should be considered anything other than a toss-up is kidding themselves. Many have argued that her likely Republican opponent, Tom Tiffany, is a uniquely weak general election candidate in his own right. But it can’t be denied that there would be a whole lot of risk involved with her as a general election candidate: The “socialism” of it all will undoubtedly be a challenge. But Francesca Hong continues to be a wild card, and she might have what it takes to play a winning hand.

Editor’s note, Aug. 5, 6:20 am ET: Kansas voters rejected a proposed amendment to change how state supreme court justices are selected. This piece below was originally published on August 4.

Four years after the Dobbs decision, abortion battles have largely taken a backseat in national news to other policy fights this cycle. Leaders in the anti-abortion movement say they’re disillusioned with President Donald Trump’s unwillingness to push for further restrictions. The federal battle over access to abortion medication seems to be settled — for now. Public awareness of that fight is limited, and even at the state level, a couple years have passed since the high-pitched but successful state-level efforts to protect abortion access in red states.

Yet a little-known battle has been playing out in Kansas, where a ballot measure to change the way state supreme court justices are selected serves as a reminder that these fights over abortion are still active, even if they take the form of creative procedural and administrative efforts. 

If it succeeds, opponents say, it would open the floodgates for similar efforts in other states to politicize the remaining independent state courts (13 other states use a method similar to Kansas) and provide a road map to undoing abortion protections in other states that have safeguarded them.

On Tuesday, Kansans are deciding whether to change the way judges on the state’s highest court are selected. Up until now, judges were selected through a nonpartisan merit-based system: When a spot opens on the bench, a nine-member commission submits a list to the governor, who then selects a replacement.

The referendum would replace this system with direct, partisan elections of judges, like any other politician. Proponents say it’s a democratic move, a way to better express the will of voters over “unelected” bureaucrats. Yet critics say it’s first and foremost about restricting abortion rights — which both the court and voters have rejected.

How Kansas became an abortion rights flashpoint

In 2019, the state supreme court recognized the right to abortion in the state constitution; three years later, after Dobbs, voters rejected a Republican-championed proposal to overturn that ruling, choosing to preserve abortion rights by a 59-41 margin in a deeply red state.

It was after that failed vote in 2022 that then-state attorney general Republican candidate Kris Kobach outlined a new strategy for undoing these protections: changing the way the supreme court is selected, electing anti-abortion candidates, and “slowly and quietly” overturning the 2019 decision and 2022 vote.

“If you look at the various proposals, one of the proposals that is crystal clear is allowing people to popularly elect justices to the supreme court, which is the most common system among all of the states in the country,” Kobach said in 2022.

He’s a bit more careful talking about this in 2026: “There is a real desire to bring back some accountability to the judiciary,” he told the New York Times this week. “It’s not just about abortion.”

Tuesday’s results would have huge ramifications, critics said, given that nonelected state courts have been a final bulwark against Republican supermajorities and governors in red states. Kansas has been an island for abortion access for women in the South and central US since the Dobbs decision.

The victory in 2022 also inspired abortion rights advocates to try to use ballot measures to enshrine these protections in about a dozen other states, which successfully overturned abortion bans in Missouri and Ohio. State courts in Utah and Wyoming, meanwhile, blocked bans.

The next frontier in the abortion wars

These independent checks on power have been under attack for years, Michael Milov-Cordoba, a court expert and counsel in the Brennan Center for Justice’s judiciary program, told me. 

“State legislators across the country are increasingly trying to undermine those courts. … That retaliation takes a few different forms, but changing how judges are selected is top of the list,” Milov-Cordoba said. “And what we’ve seen is that in many states where courts stand up for abortion rights and strike down abortion restrictions, efforts to change state judiciaries have followed.”

He noted the complete reworking of the Utah supreme court in the last year, including by expanding and packing it. Similar retooling of the state supreme court happened in Iowa in 2022. Montana is likely the next battleground, where Republican legislators are exploring various avenues to strip or limit the state supreme court’s powers. That in turn has led the ACLU to get involved to boost a constitutional amendment this year that would preserve the nonpartisan status of judicial elections.

And in recent years, state supreme court elections have been inundated with spending and attention from both in- and out-of-state backers. Last year’s Wisconsin judicial election, for example, became the most expensive race of its kind in US history, with more than $100 million spent — just two years after a contest for a different seat on the state supreme court had set its own record at more than $50 million spent.

“If these courts become partisan, then they will be just like any other election fight — subject to being bought and sold by billionaire political funders,” Deirdre Schifeling, the chief political and advocacy officer at the ACLU, told me. “They become a race to the bottom … and [Republicans] are finding backdoor ways to seize power to push their ideological agenda.”

Already, Kansas has seen heavy spending: More than $12 million have been spent from both sides, including about $5 million from the ACLU, Schifeling told me.

”Everyone’s focused on the Michigan Senate race, other kind of big-ticket, sexy elections,” Schifeling said. “Well, this is just as important and much more under the radar. Without a balanced court, we are going to lose rights and liberties across the board in a state like Kansas.”

This story appeared in The Logoff, a daily newsletter that helps you stay informed about the Trump administration without letting political news take over your life. Subscribe here.

Welcome to The Logoff: It’s clearer than ever that President Donald Trump has no way out on Iran. 

What’s happening? Over the weekend, Trump posted to social media that he had agreed to “hold off any attack” on Iran because “the perimeters of a deal has been agreed to.” The deal, he said, would include the reopening of the Strait of Hormuz and a nebulous “end to Iran’s nuclear threat.” 

That should all sound fairly familiar, since it’s essentially what the US and Iran agreed to — and subsequently abandoned — in June. As of Monday morning, though, Trump had already pivoted, accusing Iran of being “unbelievably duplicitous” in negotiating.

“This is a last chance for them to sign a good document,” he told reporters in the Oval Office Monday afternoon, threatening “decapitation” — essentially the same warning he issued in April. 

What’s the context? Since the collapse of the June “memorandum of understanding” between the US and Iran, which ended the immediate conflict and set out a path for negotiations on Iran’s nuclear program, the two countries have resumed hostilities, if not full-scale war. In July, three more US servicemembers were killed in Iranian attacks.

The strait also remains largely closed, with only a handful of vessels per day passing through.

Why does this matter? The Logoff has tried to steer clear of blow-by-blow coverage of the Iran war, because this kind of reversal happens a lot: Trump is constantly blustering, promising a deal, threatening massive retaliation, backing down, and doing it all over again. 

This latest episode, more than five months into the war and tidily compressed into just a few days, is little different. We’re writing about it anyway because it’s a good reminder of what has emerged as the essential fact of the war: Trump may have started the conflict, but he doesn’t seem to have any real plan in place to end it — and the longer things drag on, the greater the risk to the US troops fighting it.

And with that, it’s time to log off…

Vox’s Unexplainable podcast is tackling the “holy grail of sensory biology” — magnetoreception, which lets birds navigate using Earth’s magnetic field — this week with some help from author and journalist Ed Yong. You can listen here, and I also recommend Yong’s newsletter, which regularly features some spectacular bird photos.

Thanks for reading, have a great evening, and we’ll see you back here tomorrow! 

For a brief moment late last week, it looked like acting Attorney General Todd Blanche’s bid to get the word “acting” removed from his job title was in serious trouble. But that moment appears to have passed, as two key Republican senators just signaled that they will vote to move his nomination forward.

Sens. John Cornyn (R-TX) and Thom Tillis (R-NC) halted Blanche’s nomination because of concerns about a $1.8 billion slush fund that Blanche helped create as the Justice Department’s acting leader, as well as some objections to a collusive deal between President Donald Trump and the IRS that could have shut down tax investigations into Trump, his family, and his businesses. Both senators announced on Monday that they will no longer block the nomination.

Though these two senators, both of whom are retiring in January, did extract some minor concessions from the Department of Justice in return for their votes, they objected to only a small fraction of Blanche’s efforts to transform the DOJ’s lawyers into Trump’s personal advocates and enforcers. The fact that Blanche is likely to be confirmed, despite his sycophancy toward Trump, suggests that any meaningful guardrails on a weaponized Justice Department have ceased to exist.

Cornyn and Tillis both currently serve on the Senate Judiciary Committee. That committee, where Republicans currently have a two-vote majority, was originally scheduled to vote on Blanche’s nomination Thursday — but that vote would have failed if Cornyn and Tillis rejected the nominee. And, now, it’s likely to go ahead.

Cornyn and Tillis’s objections to Blanche were quite narrow in scope. They demanded a written order from Blanche “terminating the anti-weaponization fund,” a reference to the $1.8 billion slush fund that could be paid out to Trump’s allies, including participants in the January 6 attack on the US Capitol. Blanche originally established this fund in a May 18 order. On Sunday night, he tweeted out a new document terminating his May 18 order. 

Cornyn also objected to the breadth of settlement between Trump and the IRS — which could be read to permanently cut off that agency’s power to audit Trump, his companies, or many members of his family — but only sought to narrow this settlement’s scope.

For the most part, however, the two senators did not object to Blanche’s broader efforts to weaponize the Justice Department against people who’ve displeased Trump, including the use of law enforcement to cow Trump’s critics. As both deputy attorney general and as acting attorney general, Blanche oversaw several highly dubious prosecutions brought against Trump’s perceived enemies and prominent Democrats.  

A video introduced in federal court, for example, indicated that Blanche personally instructed law enforcement officers to arrest Newark’s Democratic Mayor Ras Baraka — an arrest that led a federal magistrate judge to scold the DOJ for “using the immense power of the government to pursue weak cases or to make examples without sufficient cause.” Under Blanche’s oversight, the DOJ’s also brought flimsy charges against individuals like former FBI Director James Comey and New York state Attorney General Letitia James, who investigated Trump for potential wrongdoing. (One minor exception: Tillis did previously pressure Blanche to drop an investigation into former Federal Reserve chair Jerome Powell.)

So, while Cornyn and Tillis successfully pushed Blanche to make some small concessions in order to secure their votes in the Judiciary Committee, neither senator appeared interested in a confrontation over Trump and Blanche’s broader efforts to transform the DOJ into a revenge-seeking agency. And now, it appears likely that Blanche will be confirmed — and that those efforts will continue. 

What did Blanche actually agree to do in order to get Cornyn and Tillis’s votes?

Earlier this year, Trump, the Trump Organization, and Trump’s two oldest sons sued the IRS, claiming that it failed to prevent a contractor from leaking some of Trump’s tax documents. The president originally demanded a $10 billion payout from the federal government, and the lawsuit triggered widespread concerns about a conflict of interest, because Trump was both the plaintiff and the chief executive of the government he sued.

On May 18, a little over a month after Blanche became the federal government’s top lawyer, Trump (and the other plaintiffs) entered into a “settlement agreement” that did not include a $10 billion payday for Trump but that did include several provisions favorable to Trump, his family, and his companies. 

Among other things, the settlement announced the creation of the “anti-weaponization fund,” which would distribute money to an unknown group of people who claim they were unfairly targeted by previous administrations based on decisions by five individuals appointed by Blanche. It also included a provision claiming that the IRS (and potentially other federal agencies) is “FOREVER BARRED” from pursuing any claims that “have been or could have been asserted by Plaintiffs” — open-ended language which sparked fears that a wide range of potential investigations into Trump, his family, or his companies could be shut down forever.

In response to Cornyn and Tillis’s objections to this settlement, Blanche tweeted out a brief order on Monday, stating that his own May 18 order establishing the anti-weaponization fund “is rescinded and shall have no force or effect.” According to Tillis, the DOJ also “acknowledged in a binding written order that the audit settlement is limited to the plaintiffs and the scope does not extend beyond the defendants in the lawsuit.”

So, Blanche walked back the specific order that he signed to establish the $1.8 billion slush fund, but he did not repudiate the part of the settlement that immunizes Trump, his two oldest sons, and the Trump Organization from many tax investigations. According to Tillis, the DOJ merely said that this part of the settlement does not apply to parties (including federal agencies) that were not parties to Trump’s original lawsuit.

Though this deal appeared to have been enough to secure Cornyn and Tillis’s votes, and most likely to ensure Blanche’s confirmation in a Republican Senate, it is unlikely to placate Democrats. As Sen. Adam Schiff (D-CA) tweeted Tuesday morning, Blanche’s new order “doesn’t prevent payouts to violent insurrectionists in the future,” and it “leaves in place a tax-immunity deal that could help Trump skip taxes on the $2.3 billion he made last year.”

And the question of whether the DOJ may bring politically motivated charges against figures like Baraka, Comey, or James appeared to not even be part of the deal struck between Blanche, Cornyn, and Tillis.

One reason why the two senators exacted such meager concessions may be because they didn’t have a very strong hand to play against Trump. Because Blanche is currently the Senate-confirmed deputy attorney general, he can remain head of the Justice Department, in an acting role, for as long as Trump is president. Indeed, when negotiations between Blanche and the two senators appeared to hit an impasse late last week, Trump threatened to “keep Todd as Acting A.G., and push hard to get the Anti-Weaponization Bill” if Cornyn and Tillis wouldn’t support Blanche.

But that doesn’t change the fact that Blanche’s concessions are thin gruel. His statement on the slush fund merely confirms that “there is no Fund” right now. And, under the terms of the apparent deal reached with Cornyn and Tillis, Trump, his sons, and his primary business still gain broad immunity from a wide range of tax investigations. 

So, while Cornyn and Tillis may have given Trump and Blanche a brief scare, they never put up much real resistance to Trump’s nominee.

This story appeared in Today, Explained, a daily newsletter that helps you understand the most compelling news and stories of the day. Subscribe here.

We always like a bit of good democracy news — such news being in overall short supply — and acting Attorney General Todd Blanche appeared to deliver some last night. 

Blanche is currently jockeying for the permanent AG gig, which would put him in charge of the Justice Department. But his confirmation process stalled over bipartisan objections to President Donald Trump’s “anti-weaponization fund,” a $1.8 billion reparation program for people who claimed the government had wronged them. 

On Sunday, in acknowledgement of those objections, Blanche confirmed in writing that the Justice Department had rescinded the fund and had no plans to revive it. It’s a rare reminder that the Senate’s confirmation power can still constrain the White House, even if many Republican lawmakers would rather not rock the boat. And the presidential boat was very much rocked by the dustup over Blanche’s confirmation. 

In a post on Saturday, Trump insisted that he’d get some version of the fund even over senators’ objections. And on Monday, he told reporters that he didn’t sign off on the rescission. Notably, the two Republicans who forced that surrender — North Carolina’s Thom Tillis and Texas’s John Cornyn — will both depart the Senate in early 2027.

All the president’s slush funds

The anti-weaponization fund is just the latest of Trump’s attempts to siphon federal money into an account that he or his allies control. In fact, the quest to secure such a slush fund is a defining feature of his presidency — and several of his previous, less egregious attempts have been more successful. Those include: 

1. Venezuelan oil accounts: Since the US toppled Venezuelan President Nicolás Maduro earlier this year, proceeds from the sale of Venezuelan oil have gotten routed into special US Treasury accounts, which are labeled as Venezuela’s sovereign property — but only released at the direction of the secretary of state.   
2. The Board of Peace: Trump’s controversial coalition for rebuilding Gaza was conceived as a pay-to-play operation: Member countries get seats for a mere three years unless they pony up $1 billion to the Trump-controlled organization. While more than 20 countries have signed on, Reuters reported in April that only two had actually made contributions.  
3. US sovereign wealth fund: Trump first floated this idea of a government-owned wealth fund, which would allow him to direct federal investment in private companies, during his first term. While it never got off the ground as such, the administration has since acquired equity stakes in at least a dozen firms.

“The long-term risk is not just that Trump might be doing something illegal,” Tad DeHaven, a policy analyst with the Cato Institute, wrote for Vox earlier this year. “The long-term risk is that his presidency is normalizing treating the receipt and disbursement of money as instruments of personal power.”

One link for later

➨ Happiness is overrated. In a new book, the author Ian Bogost argues we should instead pursue sensory gratification: the simple and immediate pleasure found in physical experiences, whether changing gears in a stick-shift car or holding a warm mug in your hands. 

Before you go…

• Did you know…that dolphins have been known to beat and kill other animals for fun? I spent much of my weekend watching a baby animal documentary with my sick child, and am actually devastated to learn that many adorable animals “are also murderous monsters.” 
• Today’s trivia: What Pennsylvania-born singer/songwriter is sometimes called “the godmother of punk”? (You can find this and other brain puzzles in Vox’s daily crossword. Look for the answer in tomorrow’s edition.)
• Yesterday’s trivia: Last Friday we asked you for the Vietnamese word for festival. That would be “tết,” as in the 1968 Tet Offensive — a brutal military campaign that began during Vietnamese Lunar New Year celebrations. 

New Yorkers will soon enjoy a 30 percent discount on their meat, seafood, milk, and bread — so long as they shop at a city-owned store.

Mayor Zohran Mamdani touted such bargains this week at a press conference detailing his plans to launch five public grocery stores, in a bid to lower New Yorkers’ food costs. Under the proposal, the city would own each supermarket and dictate its pricing and labor practices, while private grocery companies would handle the day-to-day operations.

Mamdani’s proposal has been getting people worked up from the moment he unveiled it on the campaign trail for his 2025 mayoral run. For many progressives, state-owned supermarkets embody one of their movement’s highest ideals — the prioritization of public needs over private profits. For conservatives, meanwhile, “Mamdani Marts” represent an attack on free enterprise (if not, the first step on the road to Soviet breadlines).

While stimulating, these grand philosophical arguments have often overshadowed more banal but pressing questions: Would public grocery stores do more to advance Mamdani’s stated goals than other things the city could do with the same money? Has New York’s mayor discovered a way to sell groceries more efficiently than Costco does — or an approach to nutritional relief more effective than simply giving poor people more cash and food?

The answer to both of these questions appears to be “no.” And that should concern more than just New Yorkers. Mamdani’s vision is already inspiring imitators, with San Francisco and Boston both exploring their own public grocery store proposals. And in national discourse, the mayor’s policy is widely understood as an experiment — one testing a bold new theory of how governments can make food more affordable.

Unfortunately, although public grocers have merit in some circumstances, the theory underlying Mamdani’s specific plan is not merely unproven but incoherent.

Grocery socialism makes more sense for small-town Republicans

Public grocery stores have historically aimed to solve a problem that New York City does not have — a total absence of private supermarkets.

In 2018, the small town of Baldwin, Florida lost its only grocer. With only 1,600 residents — and a median income of $44,271 — the community no longer generated enough business to be worth the local IGA’s while. And no other national chain rushed in to fill the gap. 

So, the municipality opened its own grocery store, the Baldwin Market.

This attracted national media attention. And not without reason. A deeply conservative small-town trying its hand at socialism was a fun story. And it turned out that Baldwin wasn’t an aberration: Other aging rural communities in red America had turned to government grocers once private supermarkets had left them in the lurch.

Progressive policy thinkers took inspiration from these examples. In the ensuing years, proposals for using public grocery stores to combat urban food deserts — impoverished neighborhoods without convenient and affordable supermarkets — began to circulate. The merits of such plans can be debated. But their logic was straightforward: Where the market fails to provide residents with healthy food options, the government should step in. 

When post-COVID inflation sowed outrage over food prices, however, some on the left reconceived the purpose of public supermarkets: In their telling, such stores weren’t merely a means of eliminating food deserts, but also a way to make groceries more affordable.

Mamdani’s plan is principally animated by the latter goal. With more than 1,100 grocery stores and 10,000 bodegas, most of New York City is well-provisioned by private food vendors. And although parts of the municipality are under-served, the mayor is not actually concentrating his stores in such areas; his planned East Harlem location lies just blocks from an Aldi, Costco, and many other grocers. 

In a report detailing its policy, the mayor’s office makes clear that its primary aim is to drive down New Yorkers’ food bills. Yet there’s little reason to think that public grocery stores are a cost-effective way to do that.

New York City probably won’t provide groceries more efficiently than Costco

Without question, a government store can make groceries more affordable for its own customers: All it has to do is charge below-market prices.

And yet, if that store manages this feat by operating at a loss, then its shoppers’ savings will come at the broader city’s expense.

That might not be too troubling, if one pictures rich taxpayers footing the bill. But for a city like New York — which faces a structural budget deficit — revenue is a scarce resource. A tax dollar spent on public grocery stores is one that can’t be spent on nutritional assistance for low-income people. And all else equal, the latter will do more to enhance affordability: If you have a limited pool of food subsidies — and want to minimize the number of New Yorkers who can’t afford groceries — then you should spend your funds on the poor, not on whoever happens to show up at your city-owned store. 

Thus, for Mamdani’s supermarkets to be cost-effective, they can’t just subsidize low prices with taxpayer money. Rather, they need to deliver groceries more efficiently than private retailers do.

Public ownership can unlock efficiencies in certain contexts. For example, America’s private health insurance system generates massive administrative redundancies. By one estimate, if the US replaced its sprawling insurance industry with a single government payer, it could cut our health system’s annual administrative costs by $500 billion. In other words, a Medicare For All program could theoretically provide more healthcare-per-dollar than the current system does, by eliminating excess bureaucracy.

But there are no comparable inefficiencies in the grocery sector. True, a government retailer could generate some savings by declining to take a profit. But margins in the industry are slim; the average profit rate among food retailers was just 2.1 percent in 2025. Therefore, Mamdani’s stores can’t finance 30 percent discounts merely by dispensing with shareholder returns. 

Meanwhile, the city actually wants its stores to be less cost-efficient than private grocers in some respects. According to its policy brief, the government grocers will pay their workers higher wages than most retailers and may also favor “local and regional suppliers,” rather than automatically contracting with whichever food producers offer the best rate. 

The city suggests that its stores will have some cost advantages: Unlike private competitors, they will pay no rent or property taxes on their facilities. But these are merely additional subsidies, not actual efficiencies: By providing free real estate to its grocers, the city is forgoing revenue that it could otherwise collect and spend on nutritional assistance or other public goods. 

It is not yet clear how much money the city is prepared to lose each year on these stores. If Mamdani’s celebrity — and heavily advertised bargains — lure large masses of people to his markets, their steeply discounted goods will quickly sell out. In that circumstance, to avoid long stretches with empty shelves, or some complex rationing system, the city would need to rapidly restock money-losing items, compounding the stores’ operating losses.  

In theory, there is one way that government stores could benefit consumers throughout a city without being especially efficient enterprises in themselves: They could force other retailers to cut prices by accepting lower profits or discovering new efficiencies. But with grocery margins already thin, squeezing private markets further could lead to closures, thereby reducing shoppers’ options. Further, Mamdani himself insists that his stores will go out of their way to avoid harming private competitors (which, in New York, consist primarily of the sorts of small businesses that Mamdani has promised to help). 

In any case, five stores won’t substantially impact pricing citywide. And so long as each government grocer operates at a loss, scaling up the program will only deepen its costs — and thus, the tradeoff between funding public supermarkets and anti-hunger programs.

Aldi for all

At present, Mamdani’s grocery store experiment looks fairly cheap. The city estimates the stores will require $70 million in capital costs. As already noted, it’s unclear precisely how large each store’s annual operating budget will be. But even if each supermarket loses many millions each year, it still won’t make that big a difference, in the context of the city’s $125 billion budget.

Still, the opportunity costs of Mamdani’s policy are real. New York City currently plans to spend $3.1 million next year on a program called Get The Good Stuff (GTGS), which essentially gives SNAP recipients up to $10 off each time they purchase fruits and vegetables from 25 participating grocery stores. With the money slated for its public supermarkets, New York could dramatically expand the program, while also investing more money into its food banks. Alternatively, the city could modestly increase cash transfers to its lowest-income residents. 

To be sure, none of that would offer much benefit to middle-class shoppers. And the mayor is quite reasonably concerned with increasing grocery affordability for the typical New Yorker. But the city can advance that aim — while actually increasing its revenue — by easing zoning and permitting rules that currently make it difficult for large, low-cost retailers like Costco and Walmart to operate in many parts of the city. Happily, to Mamdani’s great credit, his administration’s paper on public grocery stores does briefly mention the need for permitting reforms.

Maybe public grocery stores are an end in themselves

All this said, there may still be a reasonable case for Mamdani’s stores or others like them. For example, a city’s residents might come to see such retailers as a kind of attraction, public amenity, or community space, akin to a park or library.

Further, public grocers could theoretically increase the public’s faith in the government. While the grocers are unlikely to be efficient in reality, they may appear to be. After all, their low prices will be far more visible to consumers than their operating costs. If Americans come to associate the public sector with cheap, ostensibly well-run supermarkets — rather than dreary DMV lines — that could aid the left’s broader efforts to expand the government’s remit. (Although, if the public grocers end up being characterized by overcrowding and empty shelves, the policy could further erode Americans’ confidence in the public sector.)

As a means of maximizing affordability, however, Mamdani’s program makes little sense. It will mostly just transfer income from the city’s broad population to the small subset of New Yorkers who happen to live near — or work at — one of its five stores.

Where private markets fail to provide any grocery options, public stores may be the best way for municipalities to meet their residents’ nutritional needs. But in other contexts, governments can make a bigger difference by simply putting more money in poor people’s pockets.

Both Washington and Silicon Valley are in the midst of a collective freak-out over China’s recent advancements in artificial intelligence.

The latest round of consternation was triggered this month when a little-known Chinese AI startup called Moonshot released a new large language model called Kimi K3. The conventional wisdom had been that the leading AI models developed by companies like OpenAI and Anthropic were between six to 12 months ahead of their Chinese competitors. Kimi dashed those assumptions: now, analysts say American companies may be as little as two to three months behind. 

Dean Ball, a former Trump administration official now with OpenAI, warned in a bleak post on X that models like Kimi K3 could lead to a world of “full AI communism” and a “dystopian hellscape” of AI under full government control. 

Policymakers have worried for years now about China gaining an edge over the US in the AI race. Both the Donald Trump and Joe Biden administrations took steps to slow China’s AI progress, including blocking the export of the most advanced US semiconductors.  

The White House is already reportedly considering taking steps to ban “open-weight” models — models that are easier to adapt for a user’s own purposes — like Kimi K3 in the United States. The Trump administration has also accused Moonshot of using the unauthorized “distillation” of one of Anthropic’s models — basically using another model’s outputs to train itself rather than raw data — as well as gaining access to blacklisted Nvidia chips in Thailand.

But often lost in the debates about what to do about China’s accelerating AI capabilities is the question of why the US cares about this at all. Obviously, the American companies developing the latest frontier models care about maintaining their edge, but why should it matter to Americans if the chatbot in their pocket was developed in Silicon Valley or Shanghai? And perhaps even more so, why should it matter what chatbots people in Nairobi or Brussels are using? 

The concerns in the US about Chinese AI generally fall into three broad buckets: cybersecurity concerns; military and national security concerns; and human rights or democracy concerns.

For the moment, concerns about who is winning the AI race can feel a bit abstract, but as AI becomes more embedded into governments, militaries, and ordinary people’s lives, the difference will start to be felt in a much more material way at both a national and personal level. In general, there is a growing sense that it matters which of the world’s vastly different superpowers builds the technology that could transform everything. 

“People’s relationship with AI is becoming foundational to how they live their lives, so the choices people make about whose model they use and where they are physically hosted, as they share some of their most intimate secrets and ask for life advice and business guidance, and run an increasing share of their life — those are incredibly important,” said Ryan Fedasiuk, a former State Department technology adviser now at the American Enterprise Institute. “It’s a contest between the United States and China to define the operating systems through which people live and work.”

Here’s what else America loses if it loses that contest.

Chinese AI could be more vulnerable to cyberattacks 

The concerns about using Chinese AI are in some ways a repeat of the concerns over Huawei, the Chinese telecoms firm that built much of the world’s 5G internet infrastructure, but which the US government banned from operating in the United States during the first Trump administration over concerns that the Chinese government could intercept information transmitted over these networks. 

Today, the concern is that many firms are increasingly integrating Chinese AI models into their systems, both because they are often cheaper and because they are “open-weight.” (“Weights” refer to the setting an AI model uses to process a user’s inputs. “Open-weight” models make these publicly available for users to tinker with, rather than charging for access.) 

There are some indications that Americans using Chinese AI models are already vulnerable. A Booz Allen study from earlier this year tested four Chinese models commonly used by US developers and found that three of them generated software with far more “hidden vulnerabilities” that could be exploited by hackers than their US counterparts. There’s no proof that the models were doing this intentionally, but the study did find that the models were “changing their behavior depending on who the user seemed to be or what country the request referenced.”

AI can also be used to carry out cyberattacks. Although nearly all the leading models have safety protocols meant to prevent this, they’re not bulletproof. Even Anthropic’s Claude, generally considered one of the most secure models, was adapted by Chinese hackers last year to engage in cyber espionage. The open weights of the leading Chinese models could make it even easier to strip out the safety protocols. 

AI could give China a military edge

The simplest and most obvious argument for why AI matters for American national security is that it’s all too conceivable that the US and China could be at war in the years to come, and AI could be a major factor in determining who wins. 

The conflicts in Ukraine, Gaza, and Iran have shown that modern militaries are already extensively using AI for intelligence collection and targeting. Semi- or fully-autonomous drone swarms are a major component of US plans for repelling a Chinese invasion of Taiwan. Then there’s the risk of AI being used to generate new bioweapons or other dangerous threats. 

US experts believe China has pursued a “military-civil fusion” strategy, encouraging the People’s Liberation Army and Chinese defense contractors to collaborate closely with civilian technology companies and research institutions in order to gain an edge in military AI applications like intelligence analysis and drone swarms. It’s difficult to know exactly which of these capabilities China is focusing on, but procurement data suggests leading Chinese technology firms like Deepseek and Alibaba are involved in work with potential military applications. Analysts also accuse China of using outputs from US models like ChatGPT and Claude to train AI systems that could help develop China’s defense capabilities. 

And that’s just conventional weapons. The US government has alleged that Chinese labs have “continued to engage in biological activities with potential [bioweapon] applications” amid concerns that artificial intelligence could help make such weapons more sophisticated and deadly. 

China could export digital authoritarianism

Last year, it was reported that Miiloo, a fuzzy children’s plush toy with a built-in AI chatbot, would, if prompted, happily tell users Chinese Communist Party talking points like “Taiwan is an inalienable part of China.” The hubbub over Miiloo reached the US Senate floor. While it’s hard to imagine that many users were really asking Miiloo to help clear up East Asian territorial disputes, the affair illustrated much larger concerns about the dangers of letting AI models built by an authoritarian government with one of the world’s strictest censorship regimes become the global standard. 

Chinese generative AI tools are legally required to uphold the country’s “core socialist values,” according to a document published by its national cybersecurity standards committee. So it’s little surprise that DeepSeek, the Chinese chatbot that sent shockwaves through the US tech industry in 2025, politely declines to answer when you ask it what happened on June 4, 1989, in Tiananmen Square. 

It’s not just that Chinese AI could help shape the political narratives absorbed by billions around the world, at a time when US soft power is ebbing and surveys show people in many countries already now have a more positive view of China than the United States.

 The Chinese government is also increasingly integrating AI into its own censorship and surveillance apparatus, and is exporting tools like facial recognition technology to other authoritarian countries. 

The fact that under Xi Jinping, China’s government was centralizing power and becoming more, not less, authoritarian in the years leading up to the recent advances in AI are a major factor driving the mistrust in its technology. 

“I think many of the sincere arguments about the risks of these models and what China would do with them stems from the coercive authoritarian approach of China’s current leader,” said Mieke Eoyang, former US  deputy assistant secretary of defense for cyber policy. “I don’t think we would be having this conversation in the same way with someone like [China’s previous leaders] Jiang Zemin or Hu Jintao.”

It is a serious concern if models built to conform to the values and political priorities of China’s current government become the global standard. But some are skeptical of the idea that human rights and democracy should be the goal of AI competition, worrying that the damage has already been done. The premise of that idea has gotten “shakier in recent years,” says Steven Feldstein, a senior fellow at the Carnegie Endowment and author of the book The Rise of Digital Repression. Under this administration, the US has cut support for democracy and human rights programs overseas, and often allied itself with authoritarian governments. Then there’s the fact that at least one leading chatbot often seems to mimic the racist and antisemitic views of the world’s richest man who is also an ally of the current president. 

While it’s still true that Chinese AI reflects the authoritarian values and priorities of China’s leaders, Feldstein notes, “this idea that the US is standing at the forefront of protecting and advancing democracy, human rights, that we’re not sort of there to manipulate information or to push a narrative agenda that reflects the ideological preferences of its leaders, has started to fray.” 

The race to AGI 

There’s also a set of concerns around the topic of “artificial general intelligence,” the hypothetical point at which AI exceeds human capabilities and is able to improve itself. The concern, expressed by both US government commissions and senior officials in both administrations, is that China is “racing” toward AGI and that whichever country achieves it first will have a massive geopolitical advantage. This is the type of thinking behind invocations of the nuclear-era Manhattan Project to justify massive government investments in AI development. 

Chinese leaders do not appear to view AI competition this way. “The US conversation around this is much more ‘AGI-pilled’,” says Jeffrey Ding, a professor at George Washington University and expert on US-China technology competition. “The concern here is that we are very much on the brink of this explosion of more and more powerful AI that leads to it dominating everything.” Chinese leaders, on the other hand, “generally see AI as a productivity tool.”

If Chinese AI is a problem, what should we be doing about it? 

This is not just a Beltway or Silicon Valley concern. A recent Pew survey found that 43 percent Americans believe it is very important for the US to remain the leader in AI development, versus 22 percent who said it was not that important. Interestingly, the survey also found that most Americans believe China is already ahead on AI, though the expert consensus is that it’s still slightly behind. 

“We’ve gotten so used to the fact that the US has been the leading player in technological revolutions from like mobile internet to the internet era, so it’s worrying to feel we may no longer have that dominant strength,” said Selina Xu, China and AI policy lead in the office of former Google CEO Eric Schmidt. 

Even if there’s some consensus that AI competition is a priority, there’s less agreement on how to go about it. The challenge, Xu says, is “How do you manage the very concrete national security risks that come from competing with China on AI, but not turn technological competition into blanket protectionism?”

Often, the policy responses to this challenge have been contradictory. 

The Trump administration, in its first term, pioneered the policy of restricting the export of the most advanced semiconductor chips to China, but Trump undermined that policy last year by permitting Nvidia to sell its advanced H200 chips there. The move flummoxed China hawks in Washington and went against the preferences of AI developers like Anthropic, but probably had a lot to do with lobbying by chip maker Nvidia’s Jensen Huang, CEO of the world’s most valuable company. 

In some cases, the US may be inadvertently making China’s models more appealing. In June, the Trump administration placed export controls on Anthropic’s advanced Fable model. This move prompted the company to take the model down for all users and led to the first time that AI capabilities meant for the global public took a step backward.In response, French President Emmanuel Macron warned, “We will not buy any model made by [US AI] companies if from one day to the next you can just turn off the switch.” Chinese models are hardly immune from concerns about kill switches or back doors, but if both governments involved in the AI race are seen as meddling, customers may just opt for whichever one is cheaper. 

The latest flashpoint in the debate concerns the reports that the administration is considering banning open-weight models.  This prompted an open letter from dozens of leading tech companies including Nvidia and OpenAI defending access to these models as necessary for helping the US maintain AI leadership. Advocates note that open-weight models can help respond to vulnerabilities as well as create them: When a rogue OpenAI model recently hacked into the startup Hugging Face’s systems, Hugging Face’s engineers used an open-weight model developed by China’s Z.ai to analyze the attack. 

Despite the frequent comparisons, AI is not a national security competition like the early days of nuclear weapons or the space race. It’s a technology with potentially grave national security implications, that’s also used by millions of people around the world to plan their Tuesday night dinner or help with their homework. The log-in for Claude is not carried by a military officer at the president’s side. And much of the important work on developing these new technologies is being done by private tech companies, not government labs or defense contractors. 

It may be that AI capability will help determine which country has the edge in the 21st century. It may also be that the benefits of these capabilities will be shared: Chinese companies might be no less capable than their American counterparts when it comes to developing new medications or clean energy technology. 

The challenge of crafting technology to prevent a “dystopian hellscape” is to not accidentally make the existing world worse. 

Last June, in a case that effectively guaranteed one of Alabama’s Black Democratic members of Congress would be replaced by a white Republican, the Supreme Court used a single word that delighted movement conservatives and made many longtime civil rights advocates want to weep. 

In the first sentence of its decision in Allen v. Milligan (2026), most of the justices concluded that there is a tension between the federal Voting Rights Act (VRA), the landmark law that ended Jim Crow-era race discrimination in elections, and “our colorblind Constitution.” All three of the Court’s Democratic members dissented in Allen.

The word “colorblind” might not suggest, to many casual observers, the gravity of what the phrase means, how much power it has to roll back civil-rights reforms enacted in the 1960s — and the major ideological change ushered in by this Supreme Court. Proponents of colorblindness argue that any law that draws distinctions on the basis of race, for any reason whatsoever, is equally repugnant under the Constitution. Under this theory, laws like the Voting Rights Act and similar efforts to lift up historically disadvantaged racial groups are legally indistinguishable from the very kind of Jim Crow discrimination that the VRA sought to abolish. 

It’s an idea that cuts across the Roberts Court’s race decisions, including its decisions abolishing affirmative action in nearly all universities, its decisions undercutting public school integration, and its decisions dismantling the VRA. Under the colorblind thesis, the Constitution sees a modern law intended to guarantee some Black representation in Congress as precisely as intolerable as a 19th-century law forbidding Black people from voting at all. 

As Chief Justice John Roberts summarized colorblindness in a 2007 opinion, “the way to stop discrimination on the basis of race is to stop discriminating on the basis of race.”

And lest there be any doubt, the Court’s commitment to this project extends far beyond a few pithy quotes. Just over one month before Allen, in Louisiana v. Callais (2026), the Court’s Republican majority imposed colorblindness on the laws governing US elections, effectively killing a 1982 amendment to the VRA — which was signed by President Ronald Reagan — in the process. Three years earlier, in Students for Fair Admissions v. Harvard (2023), the same justices implemented colorblindness in college and university admissions, ending affirmative action in nearly every institution of higher education in the country.

Allen’s reference to “our colorblind Constitution” confirms the breadth of this project, because if the colorblind thesis is imposed by the Constitution itself, that means that judges must strike down any law or government program that takes account of race, whether in hiring, federal contracting, university admissions, elections, or the military. (Though it is notable that the Harvard decision did dodge the question of whether service academies such as West Point may take account of race in admissions, after military leaders raised concerns that a racially diverse enlisted corps would resent being commanded by a corps of officers who do not resemble them.)

This shift in the Court’s approach to race cases is likely to lead to a purge of programs intended to benefit historically disadvantaged groups. Armed with precedents like Callais and Harvard, pro-colorblindness advocacy groups have filed a wide range of lawsuits challenging scholarships for minority youth, targeting groups that promote minority-owned businesses, and even trying to prevent universities from conducting surveys of Black students.

The GOP’s hard-right turn against laws like the Voting Rights Act is a recent development. Callais, after all, neutralized a law signed by Reagan, a Republican. An earlier Supreme Court case invalidating much of the VRA, Shelby County v. Holder (2013), struck down an extension of the VRA that was signed by Republican President George W. Bush. A significant faction within the GOP has fought against the VRA since the Nixon administration, but they repeatedly lost in the two elected branches of government and only prevailed after they captured an unelected judiciary.

But the victory of the “colorblind” theory in the conservative legal movement is also the result of decades of work. It is largely a story of how a once-fringe faction within the Republican Party took over both that party and then the Supreme Court. And, now that it controls the Court, that faction seeks to lock its approach to racial questions in place forever.

The pre-Roberts Court’s approach to race

Proponents of the colorblindness thesis often trace its origins to Justice John Marshall Harlan, the only dissenter in Plessy v. Ferguson (1896), the Supreme Court’s infamous decision permitting racial segregation. “Our Constitution is color-blind,” Harlan wrote in that dissent, “and neither knows nor tolerates classes among citizens. In respect of civil rights, all citizens are equal before the law.”

Yet, while Harlan was a racial progressive by the standards of late 19th-century legal elites, his views would offend nearly all modern-day Americans. The full context of Harlan’s famous “color-blind” line suggests that he rejected segregation, at least in part, because he viewed it as unnecessary to maintain white supremacy. “The white race deems itself to be the dominant race in this country. And so it is in prestige, in achievements, in education, in wealth and in power,” Harlan began the very same paragraph that includes the line about colorblindness. “So, I doubt not, it will continue to be for all time if it remains true to its great heritage and holds fast to the principles of constitutional liberty.”

The truth is that modern legal arguments regarding the Constitution’s racial equality provisions did not fully take shape until late in the 20th century. The Court’s landmark school desegregation decision in Brown v. Board of Education (1954) spoke only of a right to racially integrated public schools: “We conclude that in the field of public education the doctrine of ‘separate but equal’ has no place.” After Brown, the Court handed down a series of unexplained orders affirming lower court decisions that struck down segregation at public beaches, municipal golf courses, parks and elsewhere. But many of these orders were just one sentence long, and they hardly answered the question of why segregation is unconstitutional outside of the public school context.

By the 1970s, legal scholars began to split between two competing theories of how the Constitution approaches race. The first theory, which Yale law professor Reva Siegel labels the “antisubordination” principle, contends that “it is wrong for the state to engage in practices that enforce the inferior social status of historically oppressed groups.” On this theory, Brown was correctly decided because the purpose of school segregation was to subordinate Black people, and the Constitution does not tolerate a racial underclass.

The other theory, which Siegel labels the “anticlassification” principle, simply declares that “government may not classify on the basis of race.” And thus any law that draws any distinction on the basis of race is constitutionally suspect, even if that law was enacted to undo the legacy of racism.

The Supreme Court of the 1970s and ’80s, however, did not divide into two neat camps. Every late 20th-century justice was at least a moderate proponent of Siegel’s anticlassification principle. While Justice Thurgood Marshall, for example, argued that the Constitution views laws which seek to “remed[y] the present effects of past racial discrimination” with less skepticism than it does laws that enforce white supremacy, he still would have subjected affirmative action and similar programs to “intermediate scrutiny,” which is the same test that courts apply to laws that discriminate against women.

Such laws are unconstitutional unless the government can provide an “exceedingly persuasive justification” for them.

Meanwhile, while a majority of the Court ultimately concluded that both Jim Crow-like laws and programs like affirmative action must survive the same exacting legal test — known as “strict scrutiny” — the Court also permitted at least some affirmative action programs to exist. The seminal case was Regents of the University of California v. Bakke (1978), which recognized that students from underrepresented racial groups bring “experiences, outlooks, and ideas” to the classroom “that enrich the training of its student body and better equip its graduates.”

And thus, if universities could not take some steps to racially diversify their student bodies, all of those students, including the white ones, would receive an inferior education.

The pre-Roberts Court, in other words, took a pragmatic approach to race cases that avoided drawing rigid lines or categorically declaring well-meaning programs intended to achieve racial equality off-limits. The justices of this era also understood that institutions outside of the Court often had far more expertise on how to foster a pluralistic society than nine lawyers in black robes.

Take Justice Lewis Powell’s opinion in Bakke, which said that “it is the business of a university to provide that atmosphere which is most conducive to speculation, experiment and creation,” and which ultimately concluded that universities “must be accorded the right to select those students who will contribute the most to the ‘robust exchange of ideas.’” 

Powell was a white Southerner who, as chair of the Richmond, Virginia, school board, left office six years after Brown with only two of Richmond’s 23,000 Black children attending school alongside white students. But he was also humble enough to recognize that universities knew something about how to educate university students to succeed in a diverse nation that he and his fellow justices did not. And so he gave these universities some leeway to exercise that expertise.

Indeed, if anything, the current Court’s most defining feature is the complete absence of the very kind of humility that Powell showed in Bakke — especially on questions of race.

Writing colorblindness into the Constitution was one of Roberts’s top priorities as chief justice

John Roberts became chief justice of the United States in 2005. Less than two years later, he made his first big push to write colorblindness into the Constitution.

In a sharp break with earlier school segregation decisions, which concluded that public schools may require each school to have “a prescribed ratio of Negro to white students reflecting the proportion for the district as a whole,” Roberts wrote that public schools are forbidden from voluntarily integrating themselves.

Parents Involved v. Seattle School District No. 1 (2007) involved two school districts, one in Seattle and another in Louisville, that allowed families to select which schools they wanted their children to attend. If more students chose a particular school than that school could accommodate, however, the districts gave a preference to children who would make their chosen school more diverse. The idea was that, since the districts had to come up with some way to determine who got to attend oversubscribed schools, it should do so in a way that fostered racial integration.

This was the case where Roberts declared that “the way to stop discrimination on the basis of race is to stop discriminating on the basis of race.” Indeed, Roberts even wrote that these two districts’ attempt to integrate public schools was no less odious than Jim Crow segregation. “Before Brown, schoolchildren were told where they could and could not go to school based on the color of their skin.” The school districts in Parents Involved, he claimed, asked his Court to “allow this once again — even for very different reasons.”

This is the essence of the colorblind theory of the Constitution. According to Roberts, the Constitution draws no distinction between white supremacist laws that segregate children by race, and an administrative scheme that tries to slightly increase the racial diversity of public schools.

It’s worth noting that Roberts’s views did not fully carry the day in Parents Involved, although a total of five justices agreed that the two school districts’ assignment schemes must be struck down. Justice Anthony Kennedy, who retired in 2018, wrote a separate opinion criticizing Roberts for going too far. “In the administration of public schools by the state and local authorities,” Kennedy concluded, “it is permissible to consider the racial makeup of schools and to adopt general policies to encourage a diverse student body, one aspect of which is its racial composition.”

But the Court’s overarching conclusion that racial integration can violate the Constitution still shocked the seniormost member of the Court. In dissent, Justice John Paul Stevens wrote that “it is my firm conviction that no Member of the Court that I joined in 1975 would have agreed with today’s decision.”

A justice ready to take colorblindness from a fringe idea into the law of the land

Stevens, however, should not have been surprised that the new chief justice would reach such a conclusion, because Roberts had spent nearly his entire career fighting for the most absolutist version of the colorblindness thesis.

In 1982, President Reagan signed an amendment to the Voting Rights Act which clarified that state election laws that have a disproportionate negative impact on racial minorities violate the VRA, even if the plaintiffs cannot prove those laws were enacted with racist intent — this was the amendment that Roberts’s Court effectively repealed in its Callais decision. Reagan did so over the objections of a conservative faction within his administration, and one of the most vocal members of this faction was a young lawyer named John Roberts.

As a Reagan White House attorney, according to journalist Ari Berman, Roberts “wrote upwards of 25 memos opposing” the 1982 law. He “prepared administration officials for their testimony before the Senate; attended weekly strategy sessions; and worked closely with like-minded senators on Capitol Hill.”

Meanwhile, in an unrelated 1982 memo to Attorney General William French Smith regarding Smith’s upcoming meeting with the chair of the US Civil Rights Commission, Roberts urged Smith to “reiterate our commitment to the color-blind principle” in that meeting.

Though Roberts, and the conservative faction that he belonged to, lost the fight to get Reagan to veto the 1982 law, this movement was hardly a flash in the pan. The original Voting Rights Act of 1965 included a provision requiring states with a history of racist election practices to “preclear” any new election laws with federal officials to ensure that those laws would not prevent people of color from voting. This provision was initially set to expire in 1970, which meant that there was an internal fight within the Nixon administration over whether President Richard Nixon should sign an extension.

The conservative faction lost that fight too. They also lost similar fights over Voting Rights Act extensions in 1975, 1982, and 2006, all under Republican presidents. The 2006 extension, signed by President Bush, passed the House by a 390-33 margin, and it passed the Senate 98-0.

Ultimately, however, a conservative faction that was too weak to convince a single senator to oppose the 2006 extension had the last laugh. While this faction was only a minor player in Bush-era electoral politics, it dominated conservative legal groups like the Federalist Society and successfully pushed its own candidates for the Supreme Court. And so, in 2013, that Court struck down preclearance in Shelby County. Roberts wrote the opinion, which was joined by all of the Court’s Republicans.

The piecemeal death of the Voting Rights Act, in other words, wasn’t just the triumph of a once-impotent faction within the GOP over a voting rights law, it was a triumph over the democratic process itself. Time and time again, this faction lost its fight against the VRA in the two democratically elected branches of government. But none of that matters to a faction that controls the Supreme Court.

The colorblindness thesis is ahistorical

One irony of this fight over how the Constitution approaches race is that, while the most recent version of the Republican Party embraces colorblindness, it does so despite a lack of evidence that the Constitution’s racial equality provisions were originally understood to support this thesis. In Republican legal circles, it’s currently fashionable to tout originalism, the idea that the Constitution’s meaning was fixed at the time it was drafted or ratified, and that judges must interpret constitutional provisions in line with this original meaning. Several of the current justices, including Justices Clarence Thomas, Neil Gorsuch, and Amy Coney Barrett, are outspoken originalists.

But within the Supreme Court it was Justice Thurgood Marshall, the great civil rights lawyer turned liberal justice, who first examined how the 14th Amendment, with its guarantee that no one will be denied the “equal protection of the laws,” was understood by the post-Civil War Americans who drafted and ratified it. And that evidence does not support colorblindness.

“The Congress that passed the Fourteenth Amendment is the same Congress that passed the 1866 Freedmen’s Bureau Act, an Act that provided many of its benefits only to Negroes,” Marshall wrote in a separate opinion accompanying the Bakke decision. He also quoted members of Congress who complained during debates over this act that it gave Black people “favors that the poor white boy in the North cannot get,” as well as lawmakers who defended the “very discrimination” the Freedmen’s Bureau “makes between ‘destitute and suffering’ negroes and destitute and suffering white paupers.”

Seven years later, in 1985, legal scholar Eric Schnapper wrote a seminal paper fleshing out Marshall’s argument. Schnapper identified “eight Reconstruction measures establishing programs limited, in varying degrees, to blacks.” These laws included the Freedmen’s Bureau named by Marshall, as well as less significant laws offering targeted benefits to Black Americans. One law appropriated $15,000 “for the relief of freedmen or destitute colored people in the District of Columbia.” Another provided safeguards for Black veterans of the Union army.

For more than four decades, no justice even attempted to rebut Marshall and Schnapper’s originalist case for affirmative action. Indeed, the Court’s originalists were so silent that, in a 2013 article, conservative legal scholar Michael Rappaport complained that they have “not made any real effort to justify their affirmative action opinions based on the Constitution’s original meaning.”

Justice Thomas did finally attempt to rebut at least some of Marshall and Schnapper’s evidence in a concurring opinion in the Harvard case, but his response to the originalist case against colorblindness was quite feeble. Thomas’s opinion did “not contend that all of the individuals who put forth and ratified the Fourteenth Amendment universally believed” in colorblindness, and he claimed that the historical record is “sparse.” 

Regarding the Freedman’s Bureau, Thomas argued that the laws authorizing it “applied to freedmen (and refugees), a formally race-neutral category, not blacks writ large,” even though “freedmen” is a category that consisted entirely of Black people. Regarding the other Reconstruction-era laws identified by Schnapper, Thomas discussed two of them.

Thomas acknowledged that “an 1866 law adopted special rules and procedures for the payment of ‘colored’ servicemen in the Union Army,” but anachronistically claimed that this law does not show that its 19th-century drafters rejected colorblindness because it may survive the strict scrutiny test that the Supreme Court devised during the 20th century. Thomas also conceded that “another law, passed in 1867, provided funds for ‘freedmen or destitute colored people’ in the District of Columbia,” but speculated that Congress “may have enacted the measure” — which again, used the words “colored people” — “not because of race, but rather to address a special problem in shantytowns in the District where blacks lived.”

All of which is a long way of saying that the colorblindness thesis is a far more recent invention than its most prominent proponents acknowledge. It was primarily devised, not by post-Civil War lawmakers seeking to undo the legacy of slavery, but by late 20th-century conservatives who opposed programs like affirmative action. 

If there was a serious originalist case for colorblindness, one of the justices would have made it in the 45-year period between Marshall’s opinion in Bakke and Thomas’s opinion in Harvard. And when Thomas finally did attempt to rebut the evidence gathered by Marshall and Schnapper, key prongs of his argument rested on anachronism and conjecture.

The fate of affirmative action and similar programs should be decided by elected officials, not by judges

In 2006, the people of Michigan voted to end affirmative action in that state’s university system. This ban sparked a far-fetched lawsuit claiming that it was somehow unconstitutional for the state to enact such a policy. As Judge Jeffrey Sutton wrote about this suit, the plaintiffs’ argument “transforms a potential virtue of affirmative action into a vice,” because “if there is one feature of affirmative-action programs that favors their constitutionality, it is that they grow out of the democratic process.” 

The question of what, if anything, should be done to racially diversify campuses — or to otherwise lift up communities that faced hundreds of years of enslavement, segregation, or other forms of government-sanctioned racism — is fraught. It is entirely reasonable to believe, as Chief Justice Roberts argued in Shelby County, that the same aggressive civil rights policies that were necessary to tear down Jim Crow in the 1960s are no longer needed two generations later. It is also reasonable to believe that, in a nation where the median white household earns over $36,000 more per year than the median Black household, that not enough is being done to close this gap.

Judge Sutton’s insight, in other words, should cut both ways. In a democracy, we are supposed to resolve these sorts of fraught questions through elections, not litigation. As the original history of the 14th Amendment shows, the United States did not ratify the post-Civil War constitutional amendments to prevent Harvard from diversifying its campus, and we certainly did not do so to prevent Congress from requiring former Confederate states to give some legislative representation to Black people. 

Similarly, proponents of colorblindness are more than capable of enacting their preferred policies without having to bypass democracy. Unlike the disenfranchised families who brought the Brown case, advocates of colorblindness do not need a judicial decree to advance their political goals — they can enact their preferred policies the same way that they did in Michigan’s 2006 election, by persuading voters that they are correct.

Nevertheless, by constitutionalizing colorblindness, the Republican justices have attempted to lock in place what could be, at most, a very transient electoral consensus around their preferred racial policy. If the American people decide, in the 2028 election or beyond, that the recent turn against laws like the Voting Rights Act was a terrible idea, they will have to replace the Supreme Court before they can reinstate those laws.

The International Criminal Court is in the trenches right now.

Last week, its chief prosecutor, Karim Khan, was voted out of his position. In a first-of-its-kind vote, 82 member states decided that he engaged in misconduct and a serious breach of duty. The vote comes after he was accused of sexual misconduct by a female colleague at the ICC, which Khan has denied.

It’s a mess, and it’s hitting the ICC at an inopportune time. On top of this internal turmoil, US Secretary of State Marco Rubio recently launched an all-out attack on the court.

In a Wall Street Journal op-ed earlier this month, Rubio wrote that the Trump administration’s goal is to “dismantle the ICC—brick by brick, if necessary,” citing the possibility that the court could “target” American citizens, such as Border Patrol agents or US soldiers.

The US is not a member of the ICC, and has a history of wariness about the organization.

Beth Van Schaack previously served as the US ambassador-at-large for global criminal justice, a role that led her to work directly with the International Criminal Court. She spoke with Today, Explained co-host Noel King about the decades-long tumultuous relationship between the US and the ICC. And they discuss whether, if the ICC is now weaker than ever, the Trump administration can actually break it for good.

Below is an excerpt of their conversation, edited for length and clarity. There’s much more in the full podcast, so listen to Today, Explained wherever you get podcasts, including Apple Podcasts, Pandora, and Spotify.

If the International Criminal Court is charged with dealing with people who have allegedly done very bad things and then trying them, why is the United States not a member?

The United States signed the treaty in the waning days of the Clinton administration, but never submitted it for ratification to the Senate. And one of the concerns is that the final treaty allows for jurisdiction over individuals who may commit crimes on the territory of state parties. And the United States has always been concerned about that principle because we have troop commitments.

Supporters of the court have always said the United States and its personnel are vulnerable regardless of whether you join. If US personnel are committing crimes on the territory of state parties, which we saw in Afghanistan, the response is to do those cases domestically and there’ll be no occasion for the International Criminal Court to consider those matters.

There’ve been peaks and valleys in the relationship between the ICC and the United States. So we’re in a frankly pretty deep valley right now with this new campaign that Secretary Rubio announced.

But back in the Bush administration, there was also very acute concern and it stemmed from this Afghanistan investigation. The Bush administration and Congress both put in place measures that they thought would be protective of US servicemembers. So that was one valley.

Now what’s interesting is in the second term of the Bush administration, many of these measures were dismantled because the Bush administration realized that the court was doing important work in some areas where it also cared about justice — in particular, Darfur.

Then, the Obama administration announced a reset in its relationship with the court, and that’s where I entered the picture. The Obama administration directed various executive agencies to look for ways to support the court where the work that it was doing was consistent with US policy.

Then, we had the first Trump administration, which reversed course almost immediately and eventually imposed sanctions on senior figures, including the prosecutor herself and one of her key aides.

Then, we get to the Biden-Harris administration. They pulled down those sanctions and again launched a reset of the relationship. At that point, there were a number of situations where the court was working, including following the brutal war of aggression by Russia against Ukraine.

Here we saw a remarkable rise of bipartisan support for the work of the court. You had senior Republican figures in Congress supporting the court. The late Sen. Lindsey Graham (R-SC) led a resolution that was supporting justice in Ukraine. Secretary Rubio, then Sen. Rubio, joined that resolution. There was new legislation passed that made it easier for the United States to support the work of the court.

Of course, with the second Trump administration, we have reversed course again, and in particular with Rubio’s announcement.

Well, let’s talk about the reversal. Secretary of State Marco Rubio is at one point on board with the ICC, and then recently he vows to dismantle it. What happened here?

It’s not clear what the precipitating event is, to be honest. There is no existential threat at present. Obviously, there’s always the concern that this is some kind of a distraction because matters are not going well in Iran and with respect to the economy — that this is something to pull eyeballs away from those matters onto something that he can look tough in. Is this just kind of red meat for anti-multilateralism or is there something that is not made public that is happening?

Is there anything actually that Marco Rubio or anyone else in the United States at the highest level could do to the International Criminal Court to put an end to it?

The International Criminal Court is part of a larger system of the rules-based order that we’ve been trying to build since World War II. Almost all of our friends and allies are members of this court and they care deeply about its success. They are committed to seeing a system of international criminal justice. So, no is the answer. The United States alone cannot dismantle this court.

Now, it can injure the court and it has already done so. There are already eight ICC judges, including judges who hail from some of our closest allies — France, for example — who have been subjected to brutal sanctions. They can’t use credit cards that pass through New York. They can’t make reservations online because of financial sanctions. They can’t travel to the United States because of travel bans, et cetera. The chief prosecutor and two of his two deputies have been sanctioned. Non-governmental organizations that are involved in human rights in Palestine have been sanctioned and a UN figure has been sanctioned. Those measures are already in place.

What his most recent campaign seems to announce is some diplomatic move to try and convince states to leave the court or to cease their cooperation with the court. And we have seen in recent days a couple of states announcing that they were withdrawing. These are mostly states within Africa, but also Venezuela, subject to investigation by the prosecutor’s office. The United States can certainly injure the court, but there’s no way that we could dismantle the court.

Most Americans don’t spend a lot of time thinking about the International Criminal Court, but they may know it because lately, it has gotten a lot of attention that the court has issued an arrest warrant for [Israeli Prime Minister] Benjamin Netanyahu. New York City Mayor Zohran Mamdani has said he would like to see Netanyahu arrested if he comes to New York.

Mamdani has said he does not have that power. He cannot order the NYPD to do it, but he would like for it to happen. What are the chances that Benjamin Netanyahu comes to the United States and is arrested?

Pretty close to zero, if not actually zero. This is a matter of federal law and international law. And the Headquarters Agreement between the United States and the United Nations, which placed the headquarters of the UN in New York, basically gives certain privileges and immunities to individuals who are coming to New York to attend official functions.

New Yorkers are a feisty bunch. So when word is out that Benjamin Netanyahu is in New York, can you imagine the protests that would happen? Can you imagine the embarrassment that he might experience when he’s supposed to be giving his remarks at the General Assembly High-level Week where all the heads of states convene to deal with transnational issues and his motorcade is ensnarled in protests protected by the First Amendment, calling for his arrest for various international crimes? If I were his lawyer, I would say, “There’s no chance you’re going to get arrested.” But honestly, it could be a miserable visit, given all of the strong support for accountability in this matter in the United States.

This is a lot of drama for a body that for a long time seems to have avoided a ton of very public drama. So who gets hurt in all of this? Is it the court? Is it the US and its reputation? Where do you think this leads?

Unfortunately, I think everyone comes out not looking great here. The court needs to put in place measures to be able to receive those complaints by witnesses, whistleblowers, victims themselves, process them quickly, but also fairly to all parties involved. The United States looks terrible as well, launching a campaign like this against an international organization whose mandate it is to deliver justice for the worst crimes known to humankind.

One of the most powerful forces in American politics is about to face a major test of its influence — and it may well come up short.

The American Israel Public Affairs Committee, or AIPAC, has for decades spent heavily to ensure that US policy remains strongly supportive of the Israeli government. The group has spent hundreds of millions of dollars on lobbying and elections to stock Congress with allies and keep them supportive of Israel’s policy priorities once they’re there. It also tries to defeat elected officials and candidates who, in its view, are too supportive of the Palestinian cause or too critical of Israel’s Gaza war.

Next week’s Democratic Senate primary in Michigan is AIPAC’s biggest-ever investment in a single election: The group has spent more than $30 million to defeat staunch Israel critic Abdul El-Sayed and to try to ensure Rep. Haley Stevens wins the Democratic nomination instead.

In the past, AIPAC has very often gotten its way. But this time may be different. El-Sayed has been explicitly running against AIPAC, arguing that the group’s heavy investment in the race discredits his opponent — and trying to capitalize on the Democratic base’s sharp turn against Israel due to the devastation of the Gaza war. And if polling is any indication, it seems to be working. Several recent surveys show El-Sayed with a double-digit lead. 

If he wins, it would arguably be AIPAC’s worst-ever defeat. It could also mean we’re in a new chapter of politics, where the group’s heavy spending is less and less effective — and even potentially toxic to voters in the Democratic base.

What has changed, and how could these changes reshape politics? 

To understand the influence of AIPAC — what it’s like to go up against the group, and also how its power may be ebbing — we spoke to a Michigan politician who ended up on their bad side, crossing the group on the Israel issue and then losing his seat. (In fact, he lost it to the very candidate AIPAC is boosting in next week’s Senate contest: Haley Stevens.)

Andy Levin was a two-term congressional incumbent and scion of a Michigan Jewish political dynasty — he succeeded his father, Sander Levin, who had served in the House for decades. Andy’s uncle, Carl Levin, had served in the Senate for decades as well; both held high-profile committee chairships.

But once in Congress, Andy Levin staked out positions on Israel that spurred furious pushback from AIPAC. A former president of the group declared he was “arguably the most corrosive member of Congress to the US-Israel relationship.”

And when, after redistricting, he and his colleague Rep. Stevens ended up running against each other in the same primary, AIPAC spent more than $4 million to boost her chances, successfully knocking Levin out of Congress.

Levin is an enthusiastic supporter of El-Sayed, and was in the car on his way to canvas for him when we spoke Thursday. Our conversation, which has been condensed and edited, follows.

How Rep. Andy Levin ran afoul of AIPAC

Tell me about your history and how you’ve thought about the Israeli-Palestinian issue over the years.

I guess the thing to understand is, I’m Jewish, right? You know what a mezuzah is? I had mezuzot on all my doors in Congress. I was the president of my synagogue until I won my primary, and then I felt I had to resign and turn that over. 

I am a long, long, longtime human rights activist. My human rights grounding comes from my values as a Jew. Everyone’s created the image of God, you know, and all this. “Justice, justice shall you pursue.” Not a lot of words are repeated twice in the Torah.

So I was active on Tibetan human rights and Haitian human rights. I was in China during the Tiananmen massacre. I interviewed the Dalai Lama. In 1990, as a graduate student, I went to Israel and Palestine. We traveled all over Israel, and we could go in Gaza, there was no Hamas yet. We met the political leaders there and talked to everybody in the West Bank.

And I came home, and I wrote an article in the Detroit Jewish News that said, “Oh my God, 23 years of occupation, this is enough. It’s totally toxic. It’s bad for our soul as the Jewish people, and there’s no solution to the Israeli-Palestinian conflict without fully realizing the political and human rights of the Palestinians.” 

My positions never really changed. But after a quarter century of Bibi [Netanyahu] being the Israeli government — he’s always been against the two-state solution, no question, but he pussyfooted around; he played a lot of games about it. But his government’s gotten more and more right-wing. And in this century, no president of either party has done anything significant to force the Israelis and the Palestinians into each other’s arms, or just to, like, really negotiate.

So you get elected to Congress in 2018. How did you earn the enmity of AIPAC?

I get to Congress. I get on the Foreign Affairs Committee. And the Gaza war before this one happens [in 2021]. It was like a pimple compared to this, but it was terrible. 

In the House, there were eight more senior Democrats who worked on Israel and Palestine together. I was, like, the kid; they let me tag along. We were all J Street-aligned [J Street is a more progressive Israel advocacy group]. 

There were these meetings, and everyone was like, we have to do reconstruction of Gaza, we have to do aid. And I’m like, okay, I’m for all that. But this is just an endless cycle of violence. I want to focus on ending the cycle of violence and actually achieving peace. 

So I wrote the Two-State Solution Act. And, dude, it was quite a milquetoast piece of legislation, if I may say so. It was not adequate in any way. But the preamble said, “whereas” — you know, all the “whereas”-es [rhetorical throat-clearing at the beginning of legislation], which aren’t an effective part of it — it said, whereas Gaza and the West Bank and East Jerusalem, under international law and United Nations resolutions, these are occupied territory.

The Democratic Majority for Israel pollster, Mark Mellman, attacked me, saying, Andy Levin says we shouldn’t be able to pray on the Western Wall. Because of that preamble, which said East Jerusalem was occupied territory arguably — it’s ridiculous.

But with the Two-State Solution Act — this is the stranglehold that AIPAC had. Some people are sort of more intellectually serious in Congress, and some are less. Some are more like, they study things themselves, but some are more staff-driven.

Some of the more serious people came up to me and said they read the Two-State Solution Act, and they said really nice things about it. And then they said, “But of course I can’t co-sponsor it, you understand.” One colleague came and co-sponsored, but because AIPAC said “dude, no,” he withdrew.

How Levin lost to Haley Stevens — who was aided by AIPAC’s spending — in 2022

So then, in the 2022 cycle, there was redistricting, and you and your colleague Haley Stevens ended up running in the same district, and that’s when AIPAC tries to defeat you.

After the Census, Michigan went from 14 to 13 seats. Haley and I were mushed in together now. You could have argued that either of us should’ve run in the new, current, 10th district. Haley lived in the 10th district; she was more moderate, and that’s a more Republican or mixed district. I never represented any of that.

My case was, I lived in the center of the new 11th district; I still live there. My kids are the fifth generation of my family to live in the new 11th district. The whole Jewish community is almost all in the new 11th district, overwhelmingly.

But the former national president of AIPAC wrote this fundraising appeal for Haley and said, Andy Levin is the most corrosive member of Congress for the US-Israel relationship. 

So the level of threat they saw in me — if you get a Rashida [Tlaib] or even if you get an Abdul [El-Sayed], they’ll fight against them. But I’m the Jewish guy, right? I’m the son of Sander Levin, the nephew of Carl Levin.

They were these two Jewish kids from Detroit, grandsons of immigrants, who rose to be, at the same time, the chair of the [House] Ways and Means Committee — my dad — and the chair of the [Senate] Armed Services Committee. So this is super threatening to AIPAC, right? So they targeted me.

So I got trounced by Haley. And I wondered afterward; I told Celinda [pollster Celinda Lake], I should have run in the new 10th district because this intramural fight was so horrible, so unpleasant. She said, I admit it would’ve been better — but I don’t admit you would’ve won in the 10th. Because AIPAC would have backed [Republican nominee] John James, and they would have spent any amount of money to beat you.

Why the politics of Israel changed — and why he thinks Abdul El-Sayed has a better shot at winning, including in the fall

You lost your race, but it certainly seems like, four years on, your views are winning inside the Democratic Party. The party base increasingly seems to be siding with you and against AIPAC now. Take me through what changed since then. 

Horrifying things caused the change. Genocide caused the change. Seventy thousand-plus Palestinians killed in Gaza caused the change. Most houses of worship, university buildings, hospitals, residential buildings being destroyed, and the US either providing the weapons for or even paying for a lot of this. 

These huge bombs, which US military leadership said, we don’t use those in populated areas — we’re providing them to Israel, and not only was Israel using them like on apartment blocks, they were using them on places they told people to flee. Any parent seeing all these maimed and killed children — and, what’s happening on the West Bank is horrifying, right? That’s part of it.

But the other part is, it gets mixed in with the larger, horrifying phenomenon of dark money, and corporate money, and special interest money in our politics. 

I mean, the idea that AIPAC is using mostly dollars from Republican centimillionaires and billionaires to be the biggest spender in Democratic primaries — that’s so gross, and it’s a huge problem for the Democratic Party and also for our democracy. 

Most people don’t vote on Israel and Palestine at all; they vote on their pocketbook, on their kids, on their neighborhood, on inflation, on jobs. We’re at a really huge gathering crisis about income and wealth inequality in America.

So we need three things, really, in candidates this year: authenticity, boldness about policy proposals, and truth-telling. I think this year, people are so fed up with politics as usual. And can you imagine? Honestly, can you imagine Abdul could win when they’re spending $80 million to beat him?

Why do you think he’s able to pull that off — if he is? The election isn’t over, of course, but it’s on Tuesday, and the latest polling has shown him with a double-digit lead. 

Sometimes the amount of money, it does become a little bit of a story in itself. Of the money being spent on Haley Stevens’s behalf, what percentage is going to be money she raised outside of AIPAC, and what percent is going to be dark money of one kind or another? 

But one of my biggest things about this election is this notion about how to win in November, which drives me out of my mind. The idea is, the liberals are gonna vote for the Democrat. The MAGA people are going to vote for the Republicans. Therefore, it’s all about the independents. It’s all about the people in the middle. Therefore, we must run to the mushy middle to win. 

It’s a static idea of the electorate, right? They call Michigan a swing state. What is a swing? The whole beauty of the swing — when you’re a kid, you go so high, you go all the way to the back, and you feel like you’re weightless for a second, and you go all the way to the front. That’s swinging. 

That’s what Michigan does. [In presidential primaries], this state voted to nominate George Wallace in 1972. We voted to nominate Jesse Jackson in 1988. And Bernie Sanders in 2016.

Abdul can attract voters across the political spectrum who hate the system. He’s so different than Trump, but they look at him and are like, well, nobody’s going to fuck with him. I don’t think he’s going to kowtow to anybody or just do what anybody says. I think he’s gonna do his own thing, and it will be entertaining to watch this. 

I think that actually Abdul can expand the electorate, and he’s the better candidate. We must win this seat in November. Abdul’s the better choice.

It’s been quite a year for the Democratic Socialists of America.

They got one of their own, Zohran Mamdani, elected mayor of New York City. They’ve taken down congressional incumbents in New York and Colorado. Their membership has soared to 120,000 dues-paying members, the group said.

But with success has come more media interest — and scrutiny.

National coverage has focused on controversial elements in the DSA’s platform, which calls for, among other things, defunding the Pentagon, taking steps toward abolishing the “carceral forces of the capitalist state,” making Jerusalem the capital of a free Palestine, and making the president of the United States chosen by Congress rather than elected. 

Internal controversies and splits within the group — such as debates over whether, and when, it should endorse a 2028 presidential candidate (such as Rep. Alexandria Ocasio-Cortez) — have also risen to national attention. Some in the movement have also argued that candidates embracing ideas like prison abolition could limit their political upside. 

As the midterm elections near, Republicans are eager to link Democrats to the DSA as part of a broader messaging strategy. President Donald Trump derided the group as communists — a term many in DSA would reject, but that some happily embrace as a self-descriptor.

So what is the DSA all about, anyway? And what do they really want?

To answer those questions, I spoke to Peter Sterne, an editor at City & State New York who has covered the ins and outs of DSA closely in such articles as “DSA for Dummies,” “Understanding DSA’s structure,” and “Get to know DSA’s internal caucuses.” 

Sterne knows the group’s inner workings well, in part, because he was a dues-paying member himself. He joined DSA in 2017 but quit it in 2023 after taking his current role covering New York politics, viewing his membership as a conflict of interest. But, he told me, he has many sources inside the group that still keep him in the loop.

Our interview, which has been condensed and edited, is below.

Why socialists aren’t just progressive Democrats

What are the core principles of DSA? What really unites the group and its members — and sets them apart from just progressive Democrats?

I would say that everyone who is in DSA identifies as a socialist, and many of them would also identify as Marxist. So they are very critical of capitalism and a society and economy that is built on capital accumulation. 

DSA members go farther than progressives like Senator [Elizabeth] Warren in that they are not just looking to rectify the excesses of capital, or create a nicer capitalism, or a mixed economy where you still have capitalism along with a generous social welfare state.

I think that, within DSA, people are critical of capitalism itself and believe in moving toward a society in which workers have more control over their workplaces and society. In terms of what that actually looks like in the short term, there are lots of different ideas. 

Now, certainly in terms of what democratic socialist candidates are running on, it is very similar to what non-socialist progressives are running on. It is largely an expansion of the social welfare state. 

But I do think there is an important philosophical difference. The Warrenites and progressive capitalists believe in the necessity of a free market economy. Whereas socialists — even if they don’t necessarily want to abolish it overnight — don’t believe that private ownership of corporations, private ownership of utilities is necessary. And that makes them much more willing to push for policies that others would see as being beyond the pale. 

The DSA’s ideological spectrum — from Bernie Sanders fans to full Communists

So then, walk me through some of the main things people within DSA disagree over — whether tactically or philosophically.

DSA identifies as a multi-tendency organization. In other words, it is a big tent of the left. 

So you have some people who are basically where Bernie is, in terms of identifying as democratic socialists — wanting to kind of bring back an FDR-style expansion of the social safety net but not necessarily wanting to go that much further. 

And then, on the other side, you would have people who identify as Communists and as Marxist-Leninists or Maoists who believe that, ultimately, the solution is a kind of Communist revolution — that, eventually, the workers are going to rise up and institute a new form of the state and society. 

They see their role as helping to radicalize the workers and prepare for that revolutionary moment, more so than trying to get elected and pass policies that will eventually move you into more of a social-welfare-type state.

So that’s the central tension. You have, on the one hand, people who would consider themselves socialists and often Marxists but believe in a democratic road to socialism, and those who identify as more revolutionary socialists, following the model of Lenin or Mao, who believe that it is impossible to reform the system.

How does that tension play out in debates inside DSA about electoral strategy?

There’s a lot of debate over red lines for elected officials and whether DSA elected officials should work within the Democratic Party or not. 

DSA generally does not just endorse people who are running for office in the way that a lot of organizations do. They believe in strategically choosing specific races to get involved in — and ideally having their own members, people who have been involved in the organization and are accountable to the organization and loyal to DSA running as candidates. 

DSA’s official strategy is that, for now, they will run candidates on the Democratic Party line. But they want to build up the infrastructure to be their own party in everything but name. So DSA wants to have its own donor networks, its own data on who is voting, its own electoral strategy — as opposed to relying on the Democratic Party for any of that.

Ideally, then, you can imagine DSA breaking from the Democratic Party. That would be the “dirty break” strategy, as opposed to the “clean break” strategy, which is, “we just stop running anyone as a Democrat right now.” The clean break strategy has been largely discredited within DSA, and they’re largely focused on the dirty break.

But as DSA wins more power running as Democrats, it becomes harder for people to imagine them ever breaking with the party. So people say they’re now doing a “dirty stay” strategy, where they’re just going to try and do realignment, which means basically taking over the Democratic Party.

The internal DSA tensions around electoral pragmatism 

There’s been some attention around how the national DSA leadership and the New York City DSA — which is seemingly the most successful of the DSA chapters around the country — have different views on these electoral strategy questions. Can you break that down?

NYC’s DSA’s political leadership is dominated by two caucuses in particular: Socialist Majority Caucus and Groundwork. Both of those are sometimes described as being on the “DSA right” —  although, they do not like that term. 

They would say that they are focused on “mass politics,” on trying to bring in the largest number of people into DSA, on trying to get DSA members elected into positions of power who can then support legislation that will materially improve the lives of the working class, can win people over to socialism and bring in more members, and build a large influential powerful socialist organization. That is essentially the Zohran theory.

Most of the people in DSA who were involved in the Mamdani campaign, and a number of them who are now involved in NYC government — such as the head of the Office of Mass Engagement, or the head of the mayor’s office to protect tenants — those are DSA members, and specifically, they are members of the Socialist Majority Caucus. SMC and Groundwork are very focused on that kind of electoral approach and, generally, are more pragmatic. They are very averse to disciplining electeds or trying to force electeds to publicly defend positions that are politically unpopular — like, say, police abolition.

Their counterparts would be the “DSA left.” Within New York City, the main left caucus is Emerge. You also have Marxist Unity Group, Red Star, and Springs of Revolution, which is primarily an anti-Zionist faction. They have less influence within New York City overall. But they have more influence within DSA nationally. A lot of chapters outside of New York City and California are more ideologically focused — and less focused on just having the most electoral success running as Democrats. 

Recently, there’s been some discussion about the national DSA platform. Francesca Hong, who’s running for governor of Wisconsin and is a democratic socialist, distanced herself from the platform this week. How should we understand the tensions around this question about appealing to voters versus being ideologically pure?

Generally speaking, I’d say most people in DSA agree on the overall ideology. The issue is more: Are you holding candidates to support that platform publicly? Do you allow your electeds to distance themselves from DSA’s positions when that seems to be the politically pragmatic choice to make? Or does that undermine the organization or even mislead the working class?

You see this a lot around anti-Zionism. There is a faction of DSA members who have been very critical of Rep. Ocasio-Cortez, because they feel that she has been insufficiently public in terms of criticizing Israel. Initially, she voted “present” on the Iron Dome. She said she was not opposed to defensive military aid to Israel. She infamously said during the DNC in 2024 that she believed that Biden and Harris were tirelessly working for a ceasefire. 

And a lot of people within DSA have been critical of AOC for what they see as compromising with Democratic leadership when it comes to Israel and Palestine and not taking a very strong kind of unconditional stance against Israel. And there have also been a lot of people, particularly within NYC DSA, who have defended AOC. Their argument is that AOC needs to be pragmatic about these things. 

You see it with Zohran in terms of his relationship with the NYPD and with NYPD commissioner Jessica Tisch. A lot of people in DSA are very critical of Tisch, who worked for [Eric] Adams. She is a billionaire and a Zionist — definitely not in keeping with DSA’s politics. And she is someone who helped to create the surveillance dragnet under the Adams administration and the De Blasio administration. 

So there have been people in NYC DSA who have called for Zohran to fire Jessica Tisch. And there are other people who say that they do not like Tisch, but they feel Zohran needs to make the pragmatic decision of working with her and not confront entrenched police power.

The DSA’s complicated relationship with Alexandria Ocasio-Cortez

Can you talk to me about the complicated relationship, the ups and downs between DSA and AOC over the years?

That relates to the tension between national and NYC. NYC DSA is really, really pro-AOC. They have a close relationship with her and work closely with her, and she really likes a lot of the leaders in NYC DSA. National DSA is much more skeptical of her, seeing her as willing to compromise and be opportunistic in ways they don’t like. 

AOC was first elected in 2018. In 2024, she was easily reendorsed for election by the NYC DSA, which then applied for the national DSA to endorse her as well. 

The national leadership dragged their feet because they were divided, particularly around her rhetoric around Israel and Palestine. 

Ultimately, they issued a conditional endorsement, saying, “we will endorse AOC if she agrees to support the boycott, divestment, and sanctions movement and oppose all aid — defensive and offensive military aid — to Israel.”

NYC DSA was concerned that AOC would not agree to that and that requiring her to agree to those conditions would alienate her from the organization. So they told national DSA, it’s fine, if you’re only going to offer a conditional endorsement we don’t need it. 

Generally, I think it’s safe to say that NYC DSA believes DSA needs AOC more than AOC needs DSA. So they’re very wary of alienating her; they want to maintain a positive working relationship. 

And all this leads into the question of what will happen in the 2028 presidential race, right? If she runs, it doesn’t seem like there’s another plausible socialist candidate in the mix.

If you talk to people who are skeptical of AOC, they would say that they would love [Rep.] Rashida Tlaib to run, because she is obviously the most committed to Palestinian solidarity. All indications are that she is not interested in running for president, so that’s kind of a dead letter. You’ll also see people say they would like Shawn Fain to run, the head of the United Auto Workers, as a real left-labor candidate harkening back to the days when union leaders were major political figures. Again, he’s very unlikely to run.

And you have some people who say that they would support Ro Khanna, even though Ro Khanna is not a socialist and is probably worse on Palestine than AOC. 

So it really does seem like it’s going to be AOC or no one. And that’s why I think that even the people who are skeptical of AOC have largely resigned themselves to the idea that if she runs, DSA will endorse. 

I think the tension now is around how supportive DSA would be. DSA could spend 2028 being an AOC surrogate, deeply involved in the campaign, making their primary goal getting her elected president in 2028. 

I think there are also people in DSA who believe that, even if AOC is the best candidate and DSA should support her, they don’t want to fully kind of liquidate themselves into the AOC campaign. They want to maintain an independent organization where DSA could set up an independent expenditure and come out with messaging that says: we support AOC but we don’t agree with her on everything. 

This story appeared in The Logoff, a daily newsletter that helps you stay informed about the Trump administration without letting political news take over your life. Subscribe here.

Welcome to The Logoff: Donald Trump’s attorney general nominee is in trouble.

What’s happening? Acting Attorney General Todd Blanche, whom Trump nominated for the permanent job last month, was set to receive a committee vote on Thursday on his way to being confirmed. Instead, lawmakers are punting on the vote after multiple key Republican senators expressed concerns — and Trump is now threatening to pull Blanche’s nomination until next year.

Why are senators worried about Blanche? Nearly all of Trump’s nominees — even his most controversial picks, like HHS Secretary Robert F. Kennedy Jr. and Defense Secretary Pete Hegseth — have been confirmed by the Senate. 

But Blanche, who is also Trump’s former personal lawyer, has a track record as acting attorney general, where he’s primarily distinguished himself by his unflinching devotion to Trump’s personal priorities: things like indicting former FBI director James Comey, creating a vast $1.8 billion slush fund to reward Trump’s allies, and immunizing the president and his family from scrutiny by the IRS via an unusual settlement deal. 

Not all of those schemes have gone well; the slush fund, for example, went down in flames in June. 

The IRS shield, however, is still kicking, and Sens. Thom Tillis (R-NC) and John Cornyn (R-TX) — both of whom sit on the Senate Judiciary Committee — are pushing Blanche to commit to changes to the settlement before they will support his nomination. So far, that hasn’t happened. 

What happens if Blanche’s nomination fails? It’s a black eye for Trump and a sign that the GOP, facing an increasingly dire midterms outlook, is growing less willing to go along with Trump’s increasingly conspicuous corruption. But a defeated vote or a pulled nomination won’t actually end Blanche’s tenure as acting attorney general — since he was previously confirmed by the Senate for the deputy attorney general job, he can likely stay in the acting role as long as Trump wants to keep him there.

And with that, it’s time to log off…

This story from the Atlantic, about a “tourism fixer” in Rome for whom just about any experience is in reach, was a great read (click here for a gift link). 

As always, have a great evening, and we’ll see you back here tomorrow!

The AI industry’s investors and critics don’t agree on much. But many in each camp share at least one basic conviction: America’s top labs are about to make a killing. 

Capital markets have signaled their faith in Anthropic and OpenAI’s impending hyper-profitability, valuing each at nearly $1 trillion. Many of Silicon Valley’s progressive adversaries also expect the labs to grow filthy rich but fear the implications, warning that AI-induced automation could transfer vast sums of money from ordinary workers to a handful of giant tech companies. Sen. Bernie Sanders’s call for nationalizing the top AI labs rests partly on that concern. 

But recent advances in Chinese AI call all of this into question.

Over the past two months, Chinese companies have released three AI models that are nearly as powerful as America’s frontier systems — and radically less expensive. 

In June, Beijing’s Z.ai debuted a model that performed nearly as well as Claude and ChatGPT’s second-tier systems on independent benchmarks. Weeks later, another Chinese firm, Moonshot, unveiled “Kimi K3,” a model that allegedly outperforms all of its American rivals except for the very latest versions of Claude and ChatGPT. Finally, just days ago, Alibaba launched a preview of Qwen3.8 Max, which purportedly outclasses even OpenAI’s most advanced systems, while trailing only Claude’s Fable in its capabilities. (Disclosure: Vox Media is one of several publishers that have signed partnership agreements with OpenAI. Our reporting remains editorially independent.)

These developments don’t merely threaten America’s AI giants with stiffer competition in the race for superintelligence. Rather, they raise a more harrowing prospect: that the AI race’s ultimate rewards will be far smaller than anticipated. In a world where new advances can regularly be leapfrogged by cheaper upstarts, hoarding the technology — and its profits — will be harder for any one company to do.

In other words, building a machine God might not be as lucrative as it’s cracked up to be. AI, it turns out, may “want to be free.”

How AI was supposed to pay off

To see how China’s new models threaten Anthropic’s profit expectations, we must first examine why those expectations have been so high.

This is not entirely self-evident. After all, AI labs aren’t much like the hyper-profitable tech giants of the 2010s. Facebook and Airbrb were relatively capital-light businesses with ultra-low marginal costs (adding a profile to Facebook or listing to Airbnb costs the companies virtually nothing). And once each gained a foothold in their respective markets, network effects enabled them to retain formidable positions without needing to constantly upgrade their products.

Building a state-of-the-art AI company is a much more involved — and astronomically more expensive — endeavor. To get to the frontier, Anthropic and OpenAI have sunk (at least) tens of billions into semiconductors, data centers, power plants, and other capital investments. Staying at the cutting-edge, meanwhile, compels them to perpetually churn out evermore costly models.

To put a new Claude model through its initial training — in which it spends months digesting the internet and sussing out statistical patterns within its text — can now cost hundreds of millions of dollars. And such foundational computation is only the beginning. A truly superlative model requires several additional months of fine-tuning. Armies of contracted experts — such as computer scientists, physicians, and mathematicians — tutor the models, grading their answers and guiding them towards better ones. Then the AI systems complete millions of rounds of practice, in which they learn through trial and error how to solve countless problems. This arduous process, known as “post-training,” compounds the costs of a single model’s development. 

All of which raises the question: Why would investors expect businesses with a cost-structure this challenging to be not merely profitable, but massively so?

There are (at least) two answers. The first (and most obvious) is that the market for superintelligent machines is liable to be vast. Frontier AI systems promise to reduce costs and improve performance in myriad white-collar sectors. And Anthropic’s soaring revenues indicate that firms do, in fact, find Claude useful. A company like AirBnB has earned billions by revolutionizing a single industry; imagine then what a technology that remade virtually all industries might be worth.

Of course, plenty of technologies are valuable but not massively profitable to produce. After all, in well-functioning markets, competition should eventually erode individual firms’ margins, even if the underlying technology continues generating huge value. 

But this is where the second answer comes in: Frontier labs’ immense costs are a burden, but they’re also a safeguard against competition — or, in industry parlance, a “moat.”

Startups may be able to afford to build or acquire more rudimentary models, many of which are “open source.” But, the thinking goes, they won’t be able to deliver Claude Fable-level performance without raising giant amounts of capital. And what investors will be willing to pour hundreds of billions into an AI pipsqueak that’s light-years behind Google, Anthropic, and OpenAI?

Alas, the Chinese AI labs’ rapid progress — and the way it was achieved — suggest that Anthropic’s moat may be shallower than previously thought.

How Moonshot swam Anthropic’s moat

The existence of powerful, Chinese AI systems is neither new nor surprising. Xi Jinping’s government has made vying for global AI dominance a key economic goal. And China’s DeepSeek, which also has stunned US companies with its lower-cost competitive models, surpassed ChatGPT as the most-downloaded free iPhone app more than a year ago.

The latest models, however, have dramatically narrowed the gap in capabilities between frontier American systems and their Chinese rivals. Just as critically, they’ve done so in a manner that other, relatively underfunded AI upstarts might be able to emulate.

Alibaba and Moonshot needed to invest massive resources to train their base models. But they allegedly found a low-cost way to refine those models into near-frontier systems: Just ask Claude.

Or, more specifically: Engage Claude in 16 million conversations, using 24,000 fake accounts. In each of those exchanges, ask the model to not only answer countless difficult questions but also, walk you through its reasoning, step by step. Then take all of this data and feed it into your own model as study material, training it to respond to the world’s most challenging queries as Claude would. 

Through this process — known as “distillation” — an AI lab can replicate virtually all of a frontier model’s capacities, without sinking vast sums into human experts and post-training computing runs. 

China’s AI labs have not admitted to using distillation. But OpenAI and Anthropic both reportedly uncovered Chinese distillation attempts earlier this year. And some of the new models appear to display tell-tale signs of distillation in conversations with ordinary users; Kimi K3 has routinely identified itself as “Claude.”

Chinese AI companies are hardly alone in using distillation to catch up with frontier labs. Earlier this year, Elon Musk admitted in court that xAI enhanced Grok’s capabilities by running distillation techniques on Claude and ChatGPT. Nonetheless, China’s latest models appear to demonstrate that distillation can help take a second-tier model to the frontier’s threshold.

America’s frontier labs have tried to defend themselves against such imitators. But this is technically difficult when distillers can assemble massive networks of bots, each asking an inconspicuous number of questions. And legally, it is difficult for America’s AI giants to argue that distillers are stealing their intellectual property. After all, in a sense, China’s copycats are merely doing to Anthropic and OpenAI what those companies did to journalists, coders, lawyers and other specialists: Feeding their public-facing outputs into a model, which then replicates their capabilities by discerning underlying patterns within the text.

Oh, and China’s giving these models away

The new Chinese models would have caused Silicon Valley enough headaches, if they merely provided stiffer competition, while demonstrating the power of distillation. 

What makes Kimi K3 and Qwen3.8 Max especially threatening to the American AI giants’ profitmaking potential, however, is that they are officially open source — meaning that the models’ parameters can be downloaded for free. (Alibaba and Moonshot have not yet released these parameters, but they say they will shortly.)

In other words, any company or hobbyist with enough computing power will soon be able to run a near-frontier Chinese model on their own hardware, modify that model to better serve a specialized purpose, and then sell access to their new version — without paying Alibiba a single yuan.

As Kimi and Qwen grow more capable, their market-share is likely to grow, at American AI giants’ expense.

For many of Anthropic and OpenAI’s potential customers, that proposition may be hard to turn down. Most businesses don’t need the world’s smartest AI, just one competent at their enterprise’s core tasks — compiling legal research, answering IT queries, writing working code, etc. A model that produces outputs 90 percent as good as Claude’s — at roughly one-sixth of the cost — will sound pretty good to many corporations.

Further, open source models aren’t just cheaper than frontier systems, but potentially more secure. If you run an AI on your firm’s own servers, then you don’t need to entrust sensitive data to Anthropic, Google, or OpenAI.

All this had led much of corporate America to embrace open-source models, even before the latest versions narrowed the capabilities gap. In a Linux Foundation survey, 63 percent of organizations reported using open-source AI systems.

And increasingly, those models are Chinese. According to Sequoia Capital, one of Silicon Valley’s premier venture capitalist firms, a majority of American AI startups now use open-source Chinese systems. As Kimi and Qwen grow more capable, their market-share is likely to grow, at American AI giants’ expense.

What’s bad for OpenAI is good (and/or catastrophic) for humanity

All this said, it is still entirely possible that OpenAI and Anthropic will justify their colossal valuations. In many highly competitive economic domains, having access to the world’s very best AI model will remain highly valuable. And America’s frontier labs still outperform all their peers. 

But it’s increasingly plausible that selling state-of-the-art AI systems will prove to be a low-margin undertaking. In a world of ubiquitous, near-frontier open source models, the AI sector’s big winners probably won’t be its top labs, but rather, its chipmakers and cloud computing providers. 

For ordinary people, a future where superintelligence is dirt cheap — and rival AI companies are constantly rising and falling, rather than consolidating into mega-corporations — would look somewhat different than the cyberpunk dystopia that the left’s been dreading. 

And not entirely in a good way. For one thing, in that reality, mitigating AI’s biggest risks would be immensely difficult. Having a handful of firms monopolize control over frontier AI systems is bad in many respects. But it does make those models easier to regulate, as the Trump administration’s decision to temporarily block Claude’s Fable in the name of cybersecurity demonstrated. 

By contrast, if recipes for ultra-powerful AI models are published all over the internet — and anyone with modest technical skills can modify them at will — then systems willing to help their users hack government bureaucracies or engineer bio-weapons are liable to proliferate.

From another angle, however, the “AI becomes almost free” scenario may look like capitalism at its finest: Retrospectively, such a development would mean that a small number of extremely rich people bankrolled the creation of an immensely useful technology, under the expectation of massive profits, only to see competition erode their returns — and disperse that tech’s benefits across a wider group of businesses and consumers. 

Granted, in the case of AI, this process might also generate a super-virus that kills us all. But hey, no system is perfect.

A court case involving President Donald Trump’s efforts to sabotage voting by mail reached the Supreme Court on Monday. The executive order at the heart of Trump v. California is one of many attempts by Trump and his Republican Party to make it harder for voters to mail their ballots and to potentially invalidate lawfully cast ballots that are submitted by mail. 

The executive order is particularly significant, because, with it, Trump is claiming the unilateral authority to change US election policy in ways that, at least according to the Constitution, can only be done by Congress or by state governments. Trump and many of his Republican allies have previously urged other institutions, such as Congress or the Supreme Court, to shift US election rules to benefit the GOP. But now, he’s claiming the power to rewrite those rules himself.

The March executive order at issue in California claims to be an effort to prevent noncitizens from voting in federal elections. But several provisions of the order seem designed to intimidate state and local election officials by threatening them with illegal prosecutions. Another provision would forbid the US Postal Service (USPS) from delivering many mailed ballots, which would have political benefits for Trump. 

Democrats have been more likely to vote by mail than Republicans since at least the Covid-era 2020 election, and this trend has continued post pandemic. So, any election policy that suppresses mailed ballots will harm Democrats and help Republicans.

In late June, a federal district court determined that several provisions of Trump’s executive order exceed his authority and must be struck down. 

The issue before the Supreme Court in California, however, does not concern whether the executive order is legal or not. Instead, Trump’s lawyers claim that the district court acted prematurely by invalidating the executive order before it was fully implemented and before several federal agencies have an opportunity to flesh out how its provisions will actually work. Those lawyers, in other words, want to buy time for federal agencies to implement parts of the order before it can be challenged in court.

This is actually a plausible argument, as Trump’s brief to the justices makes several significant concessions that, if embraced by the justices, would significantly weaken Trump’s order at least for the time being. As Trump’s lawyers now characterize his order, it doesn’t do much of anything until several federal agencies take additional steps.

Still, the stakes in this case, which arises on the Court’s “shadow docket,” a mix of emergency motions and other matters that the justices often decide very quickly, are quite high. Even if the justices do hold Trump to the promises his lawyers made in their brief and issue an opinion explicitly stating that the order currently imposes no obligations on anyone, that will still set up a future reckoning on whether Trump’s efforts to suppress voting by mail are valid. Eventually, those agencies will act, and several of them are likely to do so before the midterms, and, then, the question of whether Trump can fundamentally alter how US elections are conducted will have to be decided by the courts.

Trump’s executive order, if fully implemented, could potentially change the results of federal elections and place Republicans who did not win in office. It also matters because Trump claims the power to decide how states conduct federal elections, despite no statute that authorizes his order — thus, potentially expanding his own powers and the power of all future presidents.

So, what does Trump’s executive order do?

Trump v. California primarily concerns two sections of Trump’s executive order.

Section 2 of the order includes a thinly veiled threat that state and local election officials will face prosecution if they allow voters who aren’t pre-approved by the Trump administration to vote by mail. 

On the surface, the first part of this section seems innocuous. It requires the Department of Homeland Security “compile and transmit to the chief election official of each State a list of individuals confirmed to be United States citizens who will be above the age of 18 at the time of an upcoming Federal election and who maintain a residence in the subject State.” During a lower court proceeding in California, Trump’s Justice Department conceded that any such list would “necessarily be incomplete.” But the federal government’s decision to compile a partial list of eligible voters does not actually prevent anyone from voting.

The problem with Section 2 is that Trump’s call for DHS to compile this list is coupled with a statement that “the Attorney General shall prioritize the investigation and, as appropriate, the prosecution of State and local officials or any others involved in the administration of Federal elections who issue Federal ballots to individuals not eligible to vote in a Federal election.” 

Thus, as Judge Indira Talwani determined in her order blocking this provision, Section 2 “implicitly threatens enforcement…against election officials who furnish ballots to voters excluded” from the Trump administration’s incomplete lists. While the order does not explicitly state that an election official who mails a ballot to a lawful voter whose name does not appear on the Trump administration’s list will be prosecuted, Trump’s decision to couple his demand that the list be created with a threat of prosecution sends a clear message to election officials that they allow voters who are not on the list to vote at their own risk.

Moreover, as Talwani wrote in her opinion, one major reason why the federal government cannot compile an accurate list of all eligible voters is because the federal government’s records “do not necessarily track name changes (such as when a woman changes her name at marriage).” Women are more likely to vote for Democrats than men.

Additionally, Section 3 orders USPS to come up with new rules that require states to provide the federal government with a list of all voters who will receive a mailed ballot “no fewer than 60 days before the election,” and it forbids the post office from delivering ballots from voters whose names do not appear on the list. This section also outlines requirements for a specific kind of envelope for mail voting with certain design features — which, at the very least, creates a problem for states that have already purchased envelopes that do not comply with these new design requirements.

The Constitution gives states and Congress, not the president, the power to decide how federal elections are conducted

The Constitution provides that states, and not the federal government, get to decide who is eligible to vote in a federal election. Article I of the Constitution, for example, provides that voters who possess “the Qualifications requisite for Electors of the most numerous Branch of the State Legislature” may vote for members of the US House. And the 17th Amendment uses similar language to describe who may vote for US senators.

Thus, because each state gets to determine who may vote in its state legislative races, states also have the power to determine who can vote for members of Congress.

The Constitution’s language governing presidential elections, meanwhile, is even more explicit that states — and not Trump — get to decide who is an eligible voter. That language provides that “each State shall appoint, in such Manner as the Legislature thereof may direct,” its presidential electors.

Thus, as Talwani concluded, no federal official, including the president, has the power to determine that only “United States citizens who will be above the age of 18” may vote in a particular federal election; only states have the authority to decide who may vote for members of Congress or the president. And the federal government certainly does not have the power to compile a list of eligible voters and threaten prosecution of any state official who provides a ballot to someone not on that list.

A separate provision of the Constitution lets states determine the “Times, Places and Manner of holding Elections for Senators and Representatives,” while also permitting Congress to “make or alter such Regulations.” Thus, while the federal government cannot determine who is eligible to vote in federal elections (except through constitutional provisions such as the 15th or 19th Amendments), Congress could conceivably pass a law regulating voting by mail.

But Trump is not Congress. And, according to Judge Talwani, “no law enacted by Congress delegates authority to control mail-in voting to USPS.”

Trump’s lawyers defend his executive order by claiming that it doesn’t do anything…yet

In their brief to the justices, Trump’s Justice Department did not actually argue that the executive order is legal (although they will likely do so in a later stage of this lawsuit). Instead, they argued that Judge Talwani’s decision blocking the executive order was “premature.” 

The thrust of DOJ’s argument is that, while the executive order announces several new election-related policies, it does not actually implement any of those policies. The executive order tasks DHS with compiling the incomplete list of eligible voters, it tasks DOJ with prosecuting state officials, and it tasks the Postal Service with coming up with the specific rules governing mailed ballots.

The Constitution does not permit anyone to file a lawsuit challenging a federal policy unless they can show they were injured in some way by that policy. Trump’s lawyers argue that, until the new election policies are fully implemented, none of the California plaintiffs can know whether they will be harmed by these policies or not.

It’s a plausible argument, but largely because DOJ’s brief makes a number of concessions that strip the executive order of any real vitality, at least until it is fully implemented by federal agencies.

The brief insisted, for example, that Section 2 should not be read to authorize prosecutions against state officials who send a ballot to someone who is not on Trump’s approved list of voters. “The Order does not deem it a crime, or even evidence of a crime, for States not to use the Lists when determining voter eligibility, and it does not direct the Attorney General to bring charges on such a theory,” it read.

Similarly, it denied that any of the executive order’s new policies governing envelopes and pre-approved voters who may vote by mail must actually be implemented, at least before the postal service issues its own rules regarding Trump’s new policies. According to the brief, “the Order does not ‘direct various changes to the voting-by-mail process,’” it merely “directs the President’s Executive Branch subordinates at the Postal Service to begin a deliberative policymaking process that may ultimately include some changes.”

While at least some of these concessions take liberties with the language of the executive order, they do potentially give the justices a third option to consider before they decide the California case. The justices could leave Talwani’s decision in full effect, or they could allow Trump to implement his new restrictions on voting. But they could also block Talwani’s order while simultaneously pointing to DOJ’s concessions and indicating that they expect Trump to honor those concessions.

If the justices take this third option, that would mean that the order would impose no obligations on the states for the time being and would similarly impose no new restrictions on voters until the agencies act. Litigation over the executive order would pause, and it would resume once the order is implemented by further agency action.

The Justice Department’s concessions in this case are fairly extraordinary, and they suggest that even Trump’s own legal team recognizes that it will be easier to persuade the courts to delay a reckoning over the executive order than it will be to convince them that the order is lawful. It remains to be seen, however, whether this Supreme Court will actually enforce the Constitution in this case — and whether they will do so right away or at some future date.

The prototypical Democrat vulnerable to a primary this year has been old, in Washington a long time, and vulnerable from the left.

But in Massachusetts, there’s a race that’s scrambling that formula. Sen. Ed Markey is 80, and he’s been in Congress for 50 years. But he’s been a stalwart ally of progressives; he was, for instance, the main Senate champion of the “Green New Deal.” And the leaders of the Democrats’ left flank, like Sens. Bernie Sanders and Elizabeth Warren, are swinging into action to try and defend him. “It’s not your age, it’s the age of your ideas. And in this race, I’m the youngest guy,” Markey has said.

His opponent, Seth Moulton, a 47-year-old Marine Corps veteran, also goes against type. While “moderate” and “establishment” often are treated as synonyms, Moulton has a history of pushing back against both progressive activists and older Democratic leaders. In his current run, he’s made generational change his main issue, including some blunt warnings about Markey’s age. 

“If Democrats win the Senate, it will almost certainly be by one vote, and we can’t afford to lose that vote due to a medical absence or emergency,” Moulton told me in an interview. “That’s not ageism, that’s just actuarial tables.”

But his Senate campaign isn’t the first time he’s taken on Democratic leaders. In 2018, he led a (failed) effort to stop Nancy Pelosi from becoming speaker again, arguing again that the aging House leadership needed to clear a path for younger members. In 2024, he became one of the first prominent Democrats to bail on President Joe Biden after his disastrous debate, saying the president needed to “let new leaders rise up.”

The party establishment, Moulton told me, has “failed to be honest with the American people about the challenges we face.” He cited Democrats denying there was “an immigration problem at the southern border under Biden,” their insistence that Biden-era inflation was “transitory,” and their initial defense of Biden after the debate exposed his frailty.

Some of his criticisms of the party have rankled progressives. After President Donald Trump’s victory in 2024, he publicly criticized Democrats’ stance on trans athletes in stark terms, telling the New York Times he didn’t want his “two little girls” getting “run over on a playing field by a male or formerly male athlete” — a statement that earned him intense backlash in Massachusetts. 

Now facing a Democratic primary electorate in a very liberal state, Moulton has tried to avoid being pigeonholed as a centrist. For instance, he said he supports abolishing ICE and prosecuting many of its officials, and he said he supports legislation like the Transgender Bill of Rights. Asked about his comments on transgender youth, he told me his intention “was not to hurt or offend anybody, but to say that we need to be willing to have these tough conversations as Democrats — and not simply cede this territory to Republicans.”

Instead, his principal argument has been that Markey should pass the torch — and that frustrations with what congressional Democrats’ ineffectiveness should in part be aimed at Markey. 

“One of the first votes I’ll take in the Senate will be against Chuck Schumer for leader,” Moulton said.

Markey, for his part, has accused Moulton of throwing trans kids “under the bus” and argued his own long record and support for causes from climate to Medicare-for-all shows him to be a reliable progressive ally. 

“Congressman Moulton is trying to walk a political tightrope,” Markey campaign manager Cam Charbonnier told Vox. “He’s working to appeal to moderate voters, but that coalition alone isn’t enough to win a Massachusetts Democratic primary. So, he’s also trying to appeal to progressives, but he doesn’t have the track record to back that up. The result is a campaign that’s trying to be everything to everyone, leaving voters questioning what he actually stands for.”

Polling in the race has been sparse, but the last two public surveys both showed a surprisingly close race ; Markey led by just 5 points in one and 6 in the other. However, there were many undecided voters, and we’ll soon see whether ad campaigns — like Markey’s new one featuring Sanders, Warren, and Rep. Ayanna Pressley — shift the numbers.

If Moulton wins the September 1 primary, or even comes close, it would suggest that the Democratic base’s frustrations with gerontocracy and Washington are more than stand-ins for a debate over ideology and reflect broader concerns about older politicians losing touch, being ineffective, aging, and even dying in office.

But if Markey ends up sailing to victory, it would send a message that it can be perilous for upstart Democratic challengers like Moulton to break with the progressive base — and that validation from the Sanders wing of the party can be a very valuable thing.

Sen. Ed Markey beat back a previous primary challenge by mobilizing progressives

Though Markey entered the House of Representatives 50 years ago, his team argued that his record shows he was frequently on the right side of history — that, for instance, he started focusing on climate change as far back as the 1980s and that he supported same-sex marriage in the 1990s. (Moulton bristled at this a bit, telling me, “He was elected as a pro-life Democrat who was opposed to the racial integration of Boston schools.”)

In 2009, he and Rep. Henry Waxman co-authored Democrats’ sweeping “cap-and-trade” climate bill, which was one of President Barack Obama’s top priorities. The bill, which became colloquially known as Waxman-Markey, died in the Senate.

When Sen. John Kerry joined Obama’s Cabinet as secretary of state, Markey ran for his open Senate seat. He was the more progressive candidate in a primary against Rep. Stephen Lynch, which he won by 15 percentage points.

But toward the end of Markey’s first full Senate term, an ambitious Democrat with a famous name — then-Rep. Joe Kennedy III, grandson of RFK (and nephew of RFK Jr.) — launched a primary bid against him, also with a message of generational change.

Markey’s initial polling was quite weak, he still wasn’t very well-known in the state, and he appeared to be in serious danger. But, unexpectedly, he became an online phenomenon of sorts, as young progressives, impressed by his Senate push for the Green New Deal and early endorsement from Rep. Alexandria Ocasio-Cortez (who has not weighed in this time), rallied to his side. His campaign’s message that he was a movement champion proved successful, and he ended up winning the primary by 11 points.

When Democrats had a trifecta in Washington again under President Joe Biden, they passed their climate bill. The new law, the Inflation Reduction Act, established the Greenhouse Gas Reduction Fund, a “green bank” of sorts. The concept, in various iterations, had been one of Markey’s main causes over the years, and its enactment was a major achievement. Yet, that success was short-lived, as the Trump administration is now attempting to dismantle the program.

Rep. Seth Moulton has been a thorn in the Democratic establishment’s side

Moulton, for his part, believes he can take down Markey, because he took down a long-time Democratic incumbent once before; that’s how he got his start in politics.

Rep. John Tierney had represented his northeastern Massachusetts district for nearly two decades, when Moulton challenged him in a 2014 primary. A longtime ally of Nancy Pelosi, Tierney became vulnerable due to a gambling scandal involving his family. (His wife’s brother was accused of racketeering for operating an illegal offshore casino, and Tierney’s wife was accused of preparing false tax returns. Both were convicted and sent to prison.)

In contrast, the 35-year-old Moulton was a fresh face — a Harvard grad and a decorated combat veteran who had served four Marine Corps tours in Iraq. He proved an effective fundraiser and campaigner, upsetting Tierney by 10 percentage points.

Yet, Moulton quickly grew frustrated with Washington. After Democrats’ 2016 defeat, he became one of the party’s most outspoken critics of Pelosi. Already, by 2017, there was talk he might run for president — and ample anonymous quotes from other Democrats disparaging him as opportunistic, grasping, and disloyal.

Moulton insisted throughout that he was making a case that went beyond personal ambition. Before the 2018 midterms, he started the Serve America PAC to try and elect more members with military or national service backgrounds who were skeptical of Democratic leaders — more people like him — to Congress. But when Democrats retook the House majority, Moulton’s high-profile effort to supplant Pelosi flamed out. 

So did his subsequent presidential bid. He was in the race for four months in 2019 but dropped out when he failed to qualify for debates, warning Democrats on his way out against “veering too far left” in their choice of candidate. 

But the latest round of Democratic failures — Biden’s debate face-plant and Kamala Harris’ defeat — spurred Moulton to make another attempt at higher office. 

“Someone who has been in Congress for half a century and will be 86 years old at the end of his next term is, in fact, too old,” Moulton told the Washington Post after launching his campaign last October. “If we haven’t learned the lessons of Ruth Bader Ginsburg and Dianne Feinstein and Joe Biden…then where are we going as a party?”

Is this a race about who’s the best progressive, or about who wants to shake up the way things are done?

In addition to the close polling in recent months, Moulton’s fundraising has been impressive — aided, Markey’s allies argued, by donors associated with private equity and crypto. (A pro-Markey Super PAC has fundraised heavily from unions, as well as some investors and corporate executives.)

But when Markey and Moulton faced off on the debate stage earlier this month, nothing even close to the Joe Biden catastrophe ensued. Markey insisted he’d never been more energized, and both candidates hit their preferred messaging points. For Markey, that meant portraying himself as a dependable progressive and Moulton as an unreliable ally.

“After the 2024 election, Congressman Moulton decided that he was gonna throw trans kids under the bus — the political bus — and blame them for the loss of the 2024 election.” Markey said. “Those kids are vulnerable. Those kids need love. Those kids don’t need to be scapegoated and blamed for what was wrong with the Democratic Party.”

But in Moulton’s interview with me, he said that “it’s very hard to find any fundamental differences in the issues” between him and Markey. 

“I’m an ally for all the people and causes on all these issues that Senator Markey and I agree on, and I’m very proud of how much my voting record aligns with Senator Markey’s on the major issues,” Moulton said. “But I’m not gonna just sit back and do what the establishment tells me to do, especially when the establishment is failing.”

There was a backup plan for getting oil out of the Middle East. We may need another one.

 On July 20, the Iran-backed Yemeni rebel movement known as the Houthis declared a “blockade” against their long-time enemy Saudi Arabia, and a short time later attacked two Saudi oil tankers in the Red Sea. Shortly afterward, several ships carrying Saudi crude oil through the Red Sea made abrupt U-turns to avoid going near the Yemeni coast.  

With Iran now blocking shipping through the Strait of Hormuz, the Red Sea was supposed to be an alternative way to get oil out of the Middle East and to the world market. Now, though, the importance of the route has merely created fresh leverage in the long-simmering conflict between the Houthis and their longtime enemies in Saudi Arabia. These attacks were a major factor in pushing oil back over $100 a barrel last week.  

The Houthis have their own agenda partly unrelated to the Iran conflict. But in launching the Iran conflict, the US has inadvertently given them outsized influence in the region: The sea lane they control is more important than ever. Now, in opening up a new front, they may have just made the war harder to end. 

The Red Sea has become increasingly important to Saudi Arabia and energy markets as shippers have avoided Hormuz in recent months. Saudi Arabia has diverted around 70 percent of its oil exports via a pipeline connecting the oil fields on the Persian Gulf to the port of Yanbu on its Red Sea coast. This means that tankers need to use the Red Sea to get oil to Asia. As a result, the Red Sea now accounts for around 7 percent of the world’s energy supplies. 

This release valve is one reason why the price of oil — and the price of gas for American drivers — has not risen as much as many experts expected when the Hormuz crisis began shortly after the US-Israel strikes on Iran. 

But now, that release valve is also under threat, putting the region and the world in what shipping experts call a “double chokehold.”

The Houthis have used their location to cause chaos before: In 2023 and 2024, they attacked over 100 ships they accused of links to Israel, and dramatically drove down shipping through the Red Sea. Those attacks stopped after the Israel-Hamas ceasefire last year, but shipping has still not fully recovered. 

President Donald Trump has threatened both the Houthis and Iran with “major military punishment” over the attacks in recent days. And the attacks also make it likelier that the conflict in Iran could make oil pricier than ever, increasing political pressure on Trump. 

What do the Houthis want? 

The Houthis are a member of Iran’s network of regional proxies, known as the “Axis of Resistance,” along with Hezbollah, Hamas, and a number of Iraqi militias. But most experts believe that while they rely on Iranian support, they operate independently and don’t take their marching orders from Tehran. 

Officially known as Ansar Allah, the Houthis are members of the minority Zaydi sect of Shia Islam. They’ve been fighting for control of Yemen since the 1990s and took over Yemen’s capital city, Sanaa, in 2014, though most countries do not recognize them as Yemen’s legitimate government. 

The Houthis fought a brutal decade-long war against Yemen’s internationally recognized government and an international coalition led by Saudi Arabia, which ended with a UN-mediated ceasefire in 2022. 

The Saudi-Houthi peace has largely held since then, even as the Houthis gained global notoriety during the Gaza war. When this year’s war began, many expected the Houthis to join the fight against Iran’s enemies, the United States and Israel, but for the first few weeks of the conflict, they were conspicuously quiet. Their impact was felt in a different way, though. 

Iran’s success in using cheap drones and missiles to effectively shut down the Strait of Hormuz almost certainly drew on lessons gained from the Houthi campaign in the Red Sea in 2023 and 2024.

In March, the Houthis made their first move toward actively joining the Iran war when they launched a round of missiles at Israel. But things really escalated in mid-July when they began launching attacks on Saudi Arabia in response to an attack on the Sanaa airport, which they blamed on the Saudis. (The initial attack was claimed by Yemen’s internationally-backed government.)

In addition to the Red Sea attacks, the Houthis also claimed an attack on a Saudi oil refinery last Saturday, the group’s first attack on Saudi energy infrastructure since 2022, followed by attacks on oil depots along the Red Sea coast on Sunday. Several Iran-backed militias in Iraq also carried out their own attacks on Saudi Arabia over the past few days. 

The Houthis have now entered the conflict, but it would be a mistake to view them just as Iranian proxies. They have their own demands and agenda in this fight. 

“When the whole Iran war started, they were kind of observing and waiting for the right moment,” said Mohammed al-Basha, a US-based security analyst and Yemen expert. Basha suspects that the Houthis saw the leverage and concessions Iran had gained throughout the region and wanted to exert some leverage of their own. “Their aggressiveness went from zero to 60 last week,” he said. “They’re all in.”

All in for what, exactly? The Houthis’ demands from the Saudis include billions in reparations payments and full control of their coastline and airspace. In the long term, they aim to be Yemen’s internationally recognized government. 

The Saudis, then, who fought a brutal and unpopular war against the Houthis that ended in a stalemate, are faced with the unappealing choice between placating the group by agreeing to its demands or a return to war. 

What about the US? The US carried out more than 1,000 airstrikes over 52 days against the Houthis in 2025. While this campaign, known as Operation Rough Rider, caused significant damage to the group’s infrastructure, it did not succeed in fully restoring shipping in the Red Sea. 

Trump has threatened retaliation, but given concerns over munitions stockpiles, the US capacity to retaliate, much less expand this war to a new front, is probably limited.

How many chokeholds is too many?

For the moment, the Bab el-Mandeb is not completely closed — shipping is down around 22 percent since the blockade was declared, but 28 ships were able to transit the strait on Monday. The Houthis maintain that their blockade targets only Saudi Arabia, but Chinese supertankers have been allowed to leave the Red Sea carrying Saudi oil.  

But shippers recall that during the Houthis’ last Red Sea campaign in 2024, they were theoretically only targeting ships linked to Israel, but had a pretty broad definition of “linked to Israel.”  Under these circumstances, many shippers, and more importantly their insurers, might not want to take the risk. 

“We’ve seen what the Houthis can do,” said Noam Raydan, an expert on energy and maritime shipping at the Washington Institute for Near East Policy. “We know that they can sink ships. We know that they can hit a ship with drones, missiles, and then even board the ship. We know that they can hijack a vessel. This is why a lot of operators, in my opinion, would really be hesitant right now to go through the Bab al-Mandab.” 

There’s also still the risk that this new front in the war could escalate. Iran’s government has asked the Houthis to shut the Bab el-Mandeb entirely if the US launches airstrikes against Iran’s power network, which Trump has repeatedly threatened to do.

This adds a new complicating factor to the larger conflict in the Middle East. The fact that the price of oil never rose as high as many experts had predicted during Operation Epic Fury, the initial strikes in the spring and early summer, is probably one reason why Trump has often seemed willing to continue the conflict, rather than cutting a quick deal on Iran’s terms. But Saudi Arabia’s access to the Red Sea is one important reason why oil has not reached $120 or even $200 a barrel. A full closure of the Bab el-Mandeb would test even the ever-adaptable global oil industry’s ability to adapt.

It’s possible the Saudis may still be able to cut a deal with the Houthis that ends the attacks. But it’s also clear that for the foreseeable future, two of the Middle East’s most important energy chokepoints are going to be under continual threat of closure by adversaries of the United States. 

For now, oil can still leave the Red Sea via the Suez Canal at its northern end, then head toward Asia around the southern tip of Africa. (Since the largest tankers aren’t able to transit the Suez while fully loaded, this is a complicated process that involves them pumping their oil into a pipeline, then meeting it at the other end in the Mediterranean.) 

“It just gets harder and harder,” said Robin Mills, CEO of the Dubai-based energy analysis firm Qamar Energy. “Can they get all of the oil from Yanbu out through the Suez? I think they probably can but it’s a little bit tight. If [the Houthi campaign] were to expand, it becomes more of a problem.” 

The Suez Canal itself is also a potential target for disruption or sabotage, retired US Adm. James Stavridis recently warned. 

The disruptions at sea will increase interest in workaround projects such as a proposed pipeline from Iraq to Turkey, bypassing sea routes entirely, but these are still a long way off. The Houthis and their allies in Iraq also demonstrated this week that they can attack oil infrastructure and pipelines in addition to ships. 

The oil and shipping industries are nothing if not adaptable. Previous disruptions, whether the tanker that ran aground in Suez in 2021, the container ship pileups caused by the Covid-19 pandemic, or the Houthis’ previous Red Sea campaign, have been temporary. Goods eventually find their way to market, even if that means another month sailing around Africa.

But as the number of armed conflicts grow, and actors like the Houthis and Iran learn how to effectively weaponize the global economy’s chokepoints to punish much larger and more powerful adversaries, the stresses on the system are growing, and the question becomes whether there’s a breaking point.   

This story appeared in The Logoff, a daily newsletter that helps you stay informed about the Trump administration without letting political news take over your life. Subscribe here.

Welcome to The Logoff: The Trump administration is further limiting due process for asylum-seekers in the US.

What’s happening? On Tuesday, the administration issued a rule changing how some asylum requests are handled. Under the new policy, immigrants in more than 444,000 asylum cases could be denied the chance to speak with an asylum officer and claim asylum, which requires them to have experienced persecution, or the threat of persecution, in their home country because of their race, religion, nationality, political opinions, or another characteristic. 

Instead, those immigrants would be redirected to immigration judges — who could order their deportation without any additional process. 

The change, according to the administration, is intended to help clear a substantial backlog in the US asylum system, which has 1.4 million cases currently pending. But it fits a pattern by the Trump administration of dramatically curtailing access to asylum in the US. 

What’s the context? Under the second Trump administration, many previous immigration judges have been forced out or fired from their roles. Their replacements, as Bloomberg reported earlier this year, are minimally trained and instructed to deny most asylum claims outright. 

What’s the big picture? The Trump administration also has a long record of sending immigrants back into dangerous situations in their home countries, or even to countries to which they have no connection. 

Last year, as my colleague Ian Millhiser reported, the Supreme Court effectively allowed the administration to nullify the Convention Against Torture and send immigrants to war-torn countries like South Sudan. 

And those deportations are poised to get worse: In June, the Court also cleared the way for Trump to end temporary protected status for immigrants from Haiti and Syria, without consideration of procedural rules. Now, Immigration and Customs Enforcement is reportedly preparing to target hundreds of thousands of Haitians living in the US for deportation, even though the island nation is dealing with serious and widespread gang violence. 

With less than 100 days until the 2026 midterm elections, don’t miss my colleague Astead Herndon’s new newsletter, The Midterms, Actually. Every week, Astead breaks down the big ideas and key people behind the political shifts shaping the 2026 elections. You can subscribe here.

And with that, it’s time to log off…

Want to spend less time on social media? Here’s some excellent advice from my colleague Constance Grady, available here with a gift link. 

Thanks for reading, have a great evening, and we’ll see you back here tomorrow!

Last week, the Trump administration announced a landmark civilian nuclear cooperation agreement with Saudi Arabia. The deal would allow the kingdom to develop nuclear power using American technology and could eventually pave the way for uranium enrichment on Saudi soil. But the agreement must first undergo congressional review — and survive President Donald Trump’s social media feed.

When the Trump administration announced its agreement, it made no mention of the possibility of normalizing diplomatic relations between Saudi Arabia and Israel. But that omission and the possibility that Saudi Arabia could eventually enrich uranium alarmed Israelis — and Trump quickly threw in both conditions in a Truth Social post.

The chaos around the agreement comes as Israeli Prime Minister Benjamin Netanyahu heads to Washington, DC, this week to meet with Trump. The Saudi deal — as well as growing friction over differences on the goals and trajectory of the renewed US war with Iran and frustration in Washington over Israeli military operations in southern Lebanon — will undoubtedly be topics.

Today, Explained co-host Sean Rameswame spoke with Michael Koplow, chief policy officer at the Israel Policy Forum, about the relationship between the two countries and its leaders ahead of Netanyahu’s visit. 

Below is an excerpt of the conversation, edited for length and clarity. There’s much more in the full podcast, so listen to Today, Explained wherever you get podcasts, including Apple Podcasts, Pandora, and Spotify.

What are we expecting from the meeting between Prime Minister Netanyahu and President Trump?

I suspect this will be among the most tense meetings that they have had because there are a number of issues on the table where President Trump and Prime Minister Netanyahu are on different pages. The issue that they used to be on the same page about was Iran, and I don’t think that is the case anymore. 

As we’ve seen the Iran war unfold, the Israeli view is still the view that they had at the outset of the war, which is that this is necessary to deal with Iran’s nuclear program, with its ballistic missiles, with its proxies. And ultimately what the Israelis wanted to see, and what they still want to see, is a different regime in Iran. President Trump may still share those goals, but it’s pretty clear that, as the war has not gone the way he anticipated, he’s looking for an exit strategy and his main focus is the Strait of Hormuz, which is not something that the Israelis care about at all. 

I think at this point, when Prime Minister Netanyahu arrives at the White House, he’s probably going to try to convince the president that the original goal should be on the table, that the US and Israel should keep going, and President Trump seems to have a different view of things.

And now we have this new deal with Saudi Arabia. Was Israel taken by surprise by the deal? 

It absolutely took Israel by surprise. When the deal was announced on Wednesday, it was a deal to provide Saudi Arabia with nuclear power and nuclear infrastructure. There was no mention of Israel. There was no mention of normalization. There was no mention of anything outside of the bounds of a nuclear agreement. 

President Trump on Thursday put out his Truth Social where he said that the deal is absolutely contingent on Saudi Arabia joining the Abraham Accords. And what’s curious about that is that when the deal was announced, there was no mention of it, and the deal was actually signed. So either the president is intending to change the terms of the deal after it’s been signed, or he’s willing to back out of it if this new condition is not met. 

Or, and I suspect that this is the actual answer, he will secure some sort of commitment from the Saudis to examine joining the Abraham Accords at some point down the road when Israel has met a number of conditions and there’s been progress toward a Palestinian state. And if that is how it unfolds, it will leave the Israelis behind. And I think this comes as a pretty big shock to [Israel] given that President Trump’s previous policy was that a deal between the US and Saudi Arabia would involve normalization.

Was Netanyahu himself surprised by this deal? 

I think he must have been surprised by this deal. It is true that Trump and Netanyahu have seemed to have had a very cushy relationship, not only during this term, but during President Trump’s first term as well. But it’s not a relationship of equals.

We’ve seen President Trump — particularly over the past few months, I think in response to criticism here in the United States over the role that Israel and the role that the prime minister may have had in convincing President Trump to go to war — repeatedly talk about the fact that he is in charge, that Prime Minister Netanyahu is going to do what he says, that he is the decider and Prime Minister Netanyahu is not. 

I do think they have a good relationship, but President Trump, at least publicly, seems to be signaling that he is in charge and he wants everybody to know that. I think that Prime Minister Netanyahu undoubtedly is feeling some stress in this relationship. And when he shows up here tomorrow, he’s going to do whatever he can to try and repair any rifts that there are.

Israel wanted this war, but hates this deal that the president may very likely complete with Saudi Arabia. Did they miscalculate somewhere in here? 

I think the Israelis have miscalculated in a number of ways. And the biggest is that consistently since 2017 or 2018, you hear Israeli officials, political officials, security officials, even ordinary Israelis, talk routinely about President Trump being the best friend that Israel has ever had and exhibiting a faith in President Trump that he will never do anything that will hang Israel out to dry. 

And a lot of what they’ve done has proceeded on that assumption. And even though everybody is aware of President Trump’s transactional nature, even though everybody is aware that he is relatively volatile, that he slips on policy issues all the time, the Israelis had this faith that they would always be an exception to that rule. And I think what we’re seeing is that they’re not necessarily an exception to that rule. 

And it’s important to stress that when it comes to this Saudi nuclear deal, the problem for the Israelis is not just that it was signed without any hint of normalization between Israel and Saudi Arabia being part of the deal itself. It may open the door for countries that have signed more restrictive agreements in the region, like the UAE, to do the same. And ultimately, all of this is a bet that the trajectory of Saudi Arabia is going to continue, that it’s going to continue modernizing, continue becoming more moderate and pragmatic, and continue improving relations with Israel. 

If that’s not the case, once you’ve handed the Saudis a nuclear program, that’s irreversible. People shouldn’t forget that Iran’s nuclear program started off as a civilian nuclear program that the United States provided for it under the days of the Shah. So for Israelis, the normalization part of this is absolutely worrisome, but the deal itself is problematic too. And I think that’s where the Israelis are feeling panicked.