OMITTED

What the news leaves out.

← Omitted front page

AI Groups Pour Millions Into Midterm Campaigns

6 sources · updated 2026-10-16
Left 50% Center 33% Right 17%
3 left · 2 center · 1 right

What happened

Ahead of the 2026 U.S. congressional midterms, AI safety advocates and individual donors have spent about $48 million this election cycle on Democratic and Republican candidates who favor regulating artificial intelligence, according to a new analysis reported by Axios. That spending includes Guardrails Alliance, a tech-worker-backed group that has raised $11.7 million through a super PAC and a 501(c)(4); gifts from researchers such as OpenAI’s Juan Felipe Cerón Uribe and Redwood Research’s Aryan Bhatt; a $1 million commitment by Anthropic CEO Dario Amodei to the safety-focused group Public First; and some $40 million that Anthropic says it has given to Public First Action. A rival super PAC, Leading the Future — funded in part by $25 million from venture capitalist Marc Andreessen and $12.5 million from OpenAI President Greg Brockman — had more than $31 million on hand at the end of the second quarter and has supported more than 15 Democrats and more than 25 Republicans; Brockman later withdrew a planned second $25 million donation after an internal revolt at OpenAI. Public First Action directed $1.6 million to Democratic Rep. Valerie Foushee of North Carolina and nearly $1.5 million to Republican Texas congressional candidate Carlos De La Cruz, while Guardrails Alliance spent $412,500 against Republican Aaron Flint in a Montana House race.
Sourcing check · Partly on the record · 6 articles

2 of 5 articles cite a named source for the story. It started with unnamed sources; a named source appeared 12 hours later.

What these terms mean and how they're counted
On the record · names a source 2 of 5
On the record means the article cites a source by name for the story's main claim: a person, or an institution speaking officially, such as a company statement or a named agency's spokesperson. The opposite is an unnamed source. The count is how many of the counted articles cite at least one such source; the names are everyone named across them, each once however the articles spell it. Here: Guardrails Alliance, Juan Felipe Cerón Uribe, Dario Amodei, Marc Andreessen, Greg Brockman and Bill Gates
Primary evidence · cites a document or data 1 of 5
Primary evidence means the article cites a record you could check yourself: a court filing or ruling, a dataset or report, a recording or transcript, an official statement or press release, market data. A quote is not a document. The count is how many articles cite one. Here: campaign finance reports, Transformer
Credits another outlet 0 of 5
The article says the story comes from another outlet's reporting (“first reported by Reuters”). Those articles relay the story; they are not independent confirmation. Here: Each article presents its own reporting.
Unnamed sources 1 of 5
The main claim is attributed to people who are not identified: “people familiar with the matter”, “officials said”. The count is how many articles do this.
How it developed
When the first report landed and whether it named anyone; when the first named source and the first document appeared. Here: First report Oct 14, 09:00 UTC by Axios, with no named source. First named source 12 hours later, from New York Post. First document 12 hours later, from New York Post (campaign finance reports).
Not counted 1
Opinion, analysis and video pieces, and articles too short to assess. Every count above is out of the counted articles. An article republished under a new headline is one article.
This is about who is cited, not whether the story is true. Each source below says what it names. How we count
Omitted — what each side leaves out

Unpacked

Left-leaning coverage does not tell readers that AI safety groups spent tens of millions this cycle on candidates who favor regulation. Bloomberg instead describes industry groups spending hundreds of millions to elect people who might write the rules governing the technology. Right-leaning coverage is more complete, reporting a $48 million safety-side total and naming researchers inside the labs among the donors, along with a worker-backed safety group and a regulation-focused PAC. Left-leaning coverage leads with industry money as the force picking who will write AI rules. Right-leaning coverage opens on safety donors backing candidates "poised to crack down," and only afterward turns to the pro-innovation super PAC and its investor backers. Left-leaning coverage never names those safety PACs at all. Does one lab's check to a single PAC make up most of the safety-side total, or is that money coming from a broader donor base?
Bottom line

You can read Bloomberg on the midterms and walk away thinking only industry groups are buying AI policy. The $48 million from safety groups appears in the New York Post, not on the left.

The receipts — 2 facts one side's coverage lacks
  • OpenAI President Greg Brockman pulled a planned second $25 million donation from Leading the Future after an internal revolt at OpenAI numberin New York Post, Axios · absent from left-leaning coverage
  • Leading the Future has supported more than 15 Democrats and more than 25 Republicans numberin Axios · absent from right-leaning coverage
The Left View
Left-leaning coverage treats the midterm money as a symptom of elite control over AI. Bloomberg describes “Big AI” as pouring millions into the races, with industry groups spending hundreds of millions of dollars to elect candidates who might write the rules governing the technology. In the Guardian, Richard Wilson and Clara Long argue that warnings from tech bosses leave the core problem unsolved: “whether the humans controlling it are a tiny group of elite executives, investors and politicians vulnerable to corporate capture.” They quote OpenAI’s Sam Altman telling the UN Security Council that “if AI is to be democratic, the most important decisions cannot be made by labs in San Francisco alone,” and contend that citizens’ assemblies — unlike campaign spending — cannot be bought (“lobbyists can buy policy influence, not the outcome of a citizens’ assembly”), citing a Utah civic-lottery forum whose recommendations on datacenters, literacy, and transparency each won at least 82% support. How much AI risk society should accept is “a political one,” they write, and democracy “may be our only means of keeping its development and deployment under meaningful human control.”
The Right View
The New York Post frames the same contest as a two-front war in which “AI safety advocates, doomers, [and] pro-innovation groups” are all flooding congressional races, and it stresses the industry’s split personality: Anthropic and OpenAI are “advocating for development slowdowns even as they race to advance their own technology.” It labels safety donors “doomers,” notes gifts to candidates such as Minnesota Democrat Peggy Flanagan — who has called AI an “existential threat” to humanity — and reports that Leading the Future has used a war chest of $149 million to back pro-business candidates. The Post highlights that some safety donors say they are not driven by party so much as by a desire to compete with tech giants, that Leading the Future argues regulation will need 60 Senate votes and thus a bipartisan coalition, and that Brockman “nixed” a further donation after “internal revolt” at OpenAI.
Our Take (balanced)
The left’s strongest argument is that whoever is writing the checks — labs, investors, or “Big AI” — a small circle of executives and donors is trying to choose the people who will write AI’s rules, while the public that lives with datacenters, job disruption, and catastrophic-risk claims is not in the room; that charge draws force from the scale of super PAC money on both sides and from Altman’s own line that decisions of this magnitude “cannot be made by labs in San Francisco alone.” The right-leaning account’s strongest argument is that this is not a unified industry buying deregulation, but “safety” factions inside the same companies spending tens of millions — including Anthropic’s large gifts to Public First Action — to elect candidates who would slow or regulate AI even as those firms keep racing to build it, with both camps giving across party lines. The central unresolved tension is whether the midterm spending represents corporate capture of AI governance or an intra-elite fight between safety and acceleration camps — and whether either camp’s donors can claim to speak for democratic control of the technology.

6 sources

More stories

The week's bottom lines, in your inbox

One email a week: the five stories that mattered and what they actually mean. Free.