AI committees spent millions on the U.S. congressional midterm elections — with mixed success

Millions spent on the midterm congressional elections, with mixed results

Political groups supporting and opposing the AI industry spent millions of dollars on the 2026 U.S. midterm congressional elections, seeking to influence the future of AI regulation — but the results were mixed, TechCrunch reports. Two AI-focused committees — American Mission and Building the Future — together spent $4.8 million in support of South Carolina Senate candidate Darlene Graham, who lost. American Mission is the Republican 'super PAC' arm of the Leading the Future (LTF) group, which is backed by OpenAI president Greg Brockman and the venture firm Andreessen Horowitz. American Mission also spent more than $500,000 on each of the Republicans Aaron Flint in Montana and Laurie Buckhout in North Carolina — both lost. LTF's Democratic arm, Think Big, fared better, supporting the re-elected Florida congressman Jared Moskowitz.

Both sides: from Anthropic to 'toxic AI money'

Groups advocating stricter AI safety measures also achieved mixed results. Public First, backed by Anthropic, through an affiliated PAC supported several losing Republicans, but fared better with Democrats; it also contributed $3 million to a PAC supporting California state Senator Scott Wiener, who was leading in his race at the time of publication. Guardrails Alliance, which seeks to turn receiving money from AI companies into a political liability, spent more than $1 million against each of two losing candidates — California Congressman Jimmy Gomez and Montana's Flint — and $300,000 in support of Angela Gonzalez-Torres, who defeated Gomez.

It is hard to separate the effect from the 'democratic wave'

Separating the influence of AI spending from the overall 'democratic wave' in these elections is difficult: Elon Musk reportedly invested about half a billion dollars in supporting Republicans, and many of his candidates also lost. LTF, for its part, backed candidates it considered favorites — and also lost. Moreover, the general elections were not LTF's main focus: the group told TechCrunch that it considers primaries a more effective use of money, since there it can shape a candidate's name recognition and their AI agenda. In the primaries, LTF spent $27.38 million, including $8 million against New York Assemblymember Alex Bores, who lost his bid for a congressional seat back in June. When deciding whether to take part in general elections, LTF weighs how competitive the race is, the potential impact of spending, and whether it makes sense to save funds for future elections.

LTF's bet on primaries and Guardrails Alliance's counterargument

Guardrails Alliance political director Charlie Bletler has a different explanation for LTF's relatively modest spending in the general elections: 'They don't want their money to become a theme of these elections,' he said. As an example, he cited the Michigan state Senate primary, in which the pro-Israel lobbying group AIPAC spent nearly $30 million supporting Haley Stevens against Abdul El-Sayed; Stevens lost, and El-Sayed then won the Senate seat. 'AIPAC became the main story of that race,' Bletler noted, adding that voters are 'very distrustful and repulsed by political spending and tech-industry lobbying,' and that in a competitive race such spending would become a 'huge liability' for candidates. According to him, LTF's ads rarely touch on AI, whereas Guardrails Alliance's ads focus precisely on candidates' ties to the AI industry: 'Our goal is to make AI money politically toxic.'

The main dispute: a federal framework versus state laws

The core conflict between pro-industry and 'pro-regulation' PACs is whether Congress should establish a national framework for AI regulation, overriding state laws, or states will retain the right to introduce their own safeguards. Judging by public statements, neither side is ready to back down: LTF, which advocates a single national framework, has announced plans to spend money in the 2028 elections as well, while Guardrails Alliance intends to keep opposing candidates backed by the AI industry. For readers, this means that the question of who sets the rules for AI in the United States will remain a subject of political struggle at least until the next electoral cycle.