The artificial intelligence industry is spending heavily in the 2026 midterms, hoping to secure influence over the technology’s first generation of legislation – and New York City’s primary has emerged as the key battleground.

AI-focused Super Pacs have raised roughly $100m this cycle, of which $44m has been spent so far, in dozens of congressional races across the country. Nearly half of all spending has converged on a single Manhattan race: Tuesday’s Democratic primary in the district of NY-12.

And much of that spending has targeted a single candidate: Democratic assemblymember Alex Bores, who is running to represent New York’s 12th House district. Bores, who worked in tech before his pivot to politics, has found himself at the unlikely center of a proxy battle for the industry’s tussle for regulatory influence.

The frenzy began a year ago, when Bores sponsored the Raise Act, the second-ever US state law requiring major AI developers to publish public safety plans. By August, his congressional campaign was under siege – attack ads on TV, by text, in the mail. The effort has been funded by Think Big, an affiliate of Leading the Future, a new bipartisan network of Super Pacs created to back “pro-AI” candidates, which has poured $8.2m into the primary.

Just four donors fund its $75m war chest: venture capitalists Marc Andreessen and Ben Horowitz, and OpenAI co-founder Greg Brockman with his wife, Anna, according to data from the Federal Election Commission (FEC). Like most of Silicon Valley, the group advocates for regulating AI with a federal framework, instead of a patchwork of state laws – a compliance minefield that will hand the AI race to China, tech firms warn.

However, Leading the Future’s anti-Bores ad blitz triggered a counter-assault by a different set of Super Pacs advocating for stronger AI safeguards. They include You Can Push Back, funded by crypto billionaire Chris Larsen, and Jobs and Democracy, the Democrat-focused subsidiary of the Public First – a network of Super Pacs, founded by Brad Carson, a former Democratic congressman from Oklahoma.

The message Leading the Future was sending, in Carson’s read: regulate AI, and we will find you, wherever you are. A former Andreessen Horowitz general partner made much the same case in a New York Times op-ed last week, accusing the industry of trying to intimidate anyone who engages “too aggressively” with AI governance. Leading the Future did not respond to a request for comment.

Public First’s funding is murky, however. The dark-money group bankrolling Public First isn’t required to disclose its donors, but the artificial intelligence company Anthropic has publicly announced a $20m contribution. Since its founding, Anthropic has marketed itself as the conscience of the AI industry: a company racing to build powerful models while warning publicly about the risks they could pose, and even floating the idea of a “temporary pause” on AI development.

According to Carson, Public First has raised another $45m from various industries, including from “people who actually are currently working at the labs, from OpenAI to Google DeepMind to X.”

Combined, the tech-funded Pacs have spent $11m on the NY-12 race to counteract Leading the Future’s messaging, with ads claiming “rightwing billionaires” are trying to buy the seat, whereas Bores is “standing up to Big Tech”. It’s turned the race, as Carson put it, into “the AI civil war.”

Bores, meanwhile, has turned the primary into a referendum: “This is the first congressional race in the country where the dividing line is: can we regulate AI at all?” he says in a campaign video. Formerly considered the underdog in a competitive race, polls suggest Bores is now in a tight race with Micah Lasher, a New York assemblymember who has also campaigned in favor of AI guardrails and curbing Big Tech’s influence. “They’ve made Alex Bores into a national star,” said Carson.

Some of that may be geography as much as backlash – NY-12 leans heavily Democratic, whereas Leading the Future is led by Trump-aligned tech executives. Brookings has also named New York City the country’s most “AI-exposed” county, where a fifth of the workforce do jobs AI could plausibly take, predominantly white-collar roles like software developers, marketers and financial analysts. Brookings calls counties like it potential “hotbeds for some of the AI era’s most agitated voters”.

While Public First has positioned itself as diametrically opposed to Big Tech’s efforts to control AI policy, its industry backing may risk conflicts of interests.

“Tech companies will say ‘this needs to slow down,’ and yet either they don’t feel like they can alone make that happen or there’s not really the political will,” said Henry Ajder, a generative AI expert. Even the most cautious of executives are competing in an AI race that has created a “constant pressure to release new models quickly”, he added.

Beyond Bores, Public First has focused on supporting candidates advocating for AI advancement.

It also gave nearly $1m to the Utah congresswoman Celeste Maloy, a Republican who has pushed bipartisan legislation to crackdown on deepfakes – while lobbying for more datacenters in Utah. In Texas, it spent $1.5m million supporting House candidate Carlos De La Cruz to win the Republican nominee, who says he’s “committed to ensuring the United States wins the AI race against China” and wants to roll back green energy rules, according to his campaign site. And it gave $800,000 to Oklahoma congressman Kevin Hern – who also took money from Leading the Future, the network Public First was formed to fight.

Meanwhile, Public First has also spent big on candidates overseeing AI legislation.

The group has put $1.6m behind Representative Valerie Foushee, who co-chairs the House Democratic Commission on AI. Another co-chair, Representative Josh Gottheimer, ran a $300,000 Public First-funded ad campaign warning of AI harms. Two-thirds of the Democrats’ AI policy leadership, in other words, is now backed by a Super Pac funded primarily by Anthropic.

Among the House races seeing the most cash from both Leading the Future and Public First are those at the center of the rural datacenter roll-out. The Pacs have spent millions to elect AI-friendly candidates in primaries across Utah, Texas, Ohio, Georgia and Kentucky, despite local backlash to datacenters.

The playbook is borrowed from crypto’s 2024 run, when more than $200m in Pac money helped crypto-aligned candidates win the overwhelming majority of targeted races – including the $40m campaign that sank Sherrod Brown’s Senate bid in Ohio. But whereas AI has the money, it doesn’t have crypto’s base – during the last election, millions of investors stood to gain considerably should they elect a president who vowed to make the US “the crypto capital of the planet and the Bitcoin superpower of the world.”

Research suggests AI is politically unpopular. A recent YouGov poll found two-thirds of US voters believe it is advancing too quickly, while only one in five think its economic impact will be positive overall – views held evenly across party lines.

“The dynamics of Wall Street and the opaque sense of elites making decisions about us that don’t benefit us – I think AI companies are increasingly being seen in a similar light, whether you’re on the right or the left,” said Ajder.

On Thursday, another AI-focused Super Pac launched: Guardrails Alliance, explicitly built to counter Leading the Future. Its backers include several labor unions and Chris Hyams, the former Indeed CEO who stepped down last year over AI concerns. It won’t take corporate money, a spokesperson said.

• Will Craft and Andrew Witherspoon contributed reporting