On Wednesday, Robinhood rolled out new products that will allow people to make trades and purchases with their credit cards via agents. I don’t even need to call my cybersecurity sources; I can already hear their screams from my office. This is not investment advice, but when it comes to agentic trading, caveat emptor…

Runner-up: Following a nine-month, error-riddled run, Starbucks is sunsetting an internal A.I. tool it had deployed to automatically count inventories in its North American stores. Things might have gone better had the tool been able to consistently tell the almond and oat milks apart. Alas.

And now for the main event…

Mallory McMorrow is clear about one thing: Artificial intelligence is “massively important.” And not just in her race for Michigan’s open Senate seat, where she’s tied for the lead in the Democratic primary. “This is a technology that’s developing more rapidly than I think even those developing it had anticipated, and it could have dire consequences if policymakers wait,” she told me when we caught up earlier this week. She compared the proliferation of A.I. to that of social media, which vexed unsophisticated and Luddite lawmakers until they fumbled the chance to adequately regulate it. “Washington completely dropped the ball,” she lamented. “The stakes are much higher with A.I.”

So today, McMorrow released “People Over A.I.,” a 12-point plan intended to stake out a middle ground between Bernie Sanders and David Sacks. The document is one of the more substantial, sophisticated, and balanced political stances on A.I. we’ve seen thus far. “I hear questions about A.I. so much more regularly now than I did a year ago, so this is clearly happening incredibly quickly, and policymakers can’t afford to wait,” McMorrow said. “I hope this spurs a conversation, and that we encourage our policymakers to act.”

The plan calls for safety testing on predeployment models, transparency disclosures and safety reporting from developers, the construction of a national A.I. lab to enable independent model testing within the government, and a general push against efforts to preempt state A.I. laws. It also calls for more-stringent global export controls, intended specifically to ensure the U.S.’s edge over China, and for the establishment of global rules on A.I.—including a “binding prohibition” against “putting A.I. inside nuclear weapons decisions.” The plan ends with a series of efforts to “keep humans in charge” by establishing penalties for the use of deepfakes without consent, a right to know when A.I. is being used in decision-making (for jobs, loans, healthcare, etcetera), a right to challenge those results with a real-life arbitrator, and a law that A.I. cannot “decide on its own to kill a human being.”

McMorrow cast a wide net in writing the plan, talking to developers, employees of the major labs, members of the A.I. safety community, researchers, and constituents. Indeed, there seems to be a little something for everyone in this plan, which builds on a smaller, more focused manifesto that McMorrow published on Wednesday, intended to “put A.I. on workers’ terms.” That piece proposes establishing a workforce reinvestment fund, federal apprenticeship and retraining program, and stronger safety net for the expected rocky transition; boosting unemployment insurance, TANF, SNAP, and Medicaid “for all workers”; and levying a tax on A.I. tokens to fund it all.

As far as political considerations go, it’s quite a safe bet. “For someone who’s running statewide in Michigan, I think this is exactly the right approach, and a calculus she’s probably making that she might take on some A.I. PAC money. But she’ll pick up a heck of a lot of voters, or at least pique their interest,” Rishi Bharwani, the U.S. director for Reset Tech, a nonprofit research organization, told me. “It’s the right thing to do.”

McMorrow, who has previously proposed plans to tackle tech-adjacent issues ranging from surveillance to online child safety, is planting her flag on A.I. as both parties struggle to define the issue. From California’s Gavin Newsom to Colorado’s Jared Polis and New York’s Kathy Hochul, many governors are intent on trying to protect people while also allowing the industry to thrive. Sen. Bernie Sanders and Rep. Alexandria Ocasio-Cortez, meanwhile, want to impose a full moratorium on data centers—“a reasonable pause to the development of A.I. to ensure the safety of humanity,” in their words. And you’ve got Sen. Elizabeth Warren, who’s now interested in “taxing” the technology—a proposal that might find an unlikely ally in OpenAI’s Chris Lehane. Then there are the countless, lesser-known state legislative efforts—most notably from New York, California, and now Illinois—all while local data center bans are becoming more and more popular.

Voter concerns are palpable but also somewhat inchoate, manifesting in everything from data center NIMBYism to attempted violent attacks against OpenAI. The unsettled mood is perhaps best encapsulated by one anonymous commenter on McMorrow’s Wednesday plan, who wrote, “I appreciate your work and thoughtfulness on this but … why on earth would we do ALL OF THIS instead of saying one word, ‘No.’ No to data centers and no to technology that ruins everyone’s lives so the tech bros can be even richer and more powerful… No.”

But McMorrow is convinced that just saying “no” is insufficient and entirely unsophisticated. “My sense is that people see positive use cases for it, but that humans have to be at the center of it,” she said. “Right now you see some of the A.I. companies, C.E.O.s, and tech billionaires flat-out telling us there’s a world in which 25, 50 percent of jobs could be done by A.I.—in a very rapid period of time—and they’re just kind of saying, ‘Trust us on it.’ That doesn’t leave a lot of confidence with people on the ground.” Government, in her view, ought to establish guardrails to ensure the technology “is developed for people and not against people.”

On some level, her platform is not entirely dissimilar to that of Alex Bores, the NY-12 House candidate who worked at Palantir and has made sensible A.I. regulations a cornerstone of his campaign. Like Bores, McMorrow has probably painted a super PAC target on her back, though she professes not to be worried. “What I know is that the industry is so far at odds with where most people are,” she told me. “People get this is happening really fast, and there’s something that feels very off about it. If I am where most people are, then it doesn’t matter how much money is spent. We’ve got to be willing to stand up for people.”

She seems to have caught neither the interest nor ire of the major A.I. PACs so far; sources at both Leading the Future and Public First told me they haven’t gotten involved in the Michigan Senate race. But even if they do—and even if Michigan becomes a secondary battleground in the war for A.I. regulation—Bharwani thinks there’s a good chance it’ll only help McMorrow raise her profile ahead of the August primary. Bores, he said, is beginning to demonstrate that PAC attacks could well be “offset by a clear demonstration of values and illustration to voters that you’re going to Washington to fight for their interests.” And taking on corporate power is rarely unpopular when elections roll around. “Folks like this send a really clear signal that they’re going to go to Washington to shake things up,” Bharwani said.

As Stanford political economist Andy Hall wrote recently, we’re in a stage of clear voter anxiety about A.I., if not backlash. Sentiment has been turning more negative, but A.I. is not yet one of the major issues—like cost of living, inflation, healthcare, and crime—that Americans vote on. Hall posited that if unemployment spikes significantly in a way that is clearly tied to A.I., then we’ll see a shift to “real backlash.” Until that point, policy proposals likely won’t be radical because they won’t need to be.

Still, Bharwani thinks A.I. will be a top five issue come 2028, and is curious whether McMorrow’s efforts will encourage other candidates. “Democrats are obviously largely leaderless,” he said. “This unique constellation of Democratic leaders taking bold positions on A.I. is the making of a 2028 primary field, and the future of a party that’s much more values-forward in how they approach tech regulation.” The rightward shift of Big Tech C.E.O.s like Elon Musk and Mark Zuckerberg, Bharwani said, was “freeing for a new crop of Democratic candidates to really speak to the public interest on this.” We’ll see what happens in November.

That’s all for today. I’ll see you next week.

Ian

Need help? Review our FAQ page or contact us for assistance. For brand partnerships, email ads@puck.news.

You received this email because you signed up to receive emails from Puck, or as part of your Puck account associated with {{customer.email}}. To stop receiving this newsletter and/or manage all your email preferences, click here.

Puck is published by Heat Media LLC. 107 Greenwich St., New York, NY 10006