AI Safety Candidates Win Key Midterm Races After $110 Million Advocacy Push

Candidates who ran on stricter AI oversight won several competitive congressional races in Tuesday's midterms, backed by a $110 million political campaign that included more than $63 million from Public First, a group tied to Anthropic co-founder Dario Amodei. The results give AI safety advocates a meaningful foothold in the next Congress — and set up 2027 as the year federal AI legislation moves from discussion to votes.

What happened: Bloomberg reports that Public First, a political organization backed by Amodei, spent more than $63 million of a combined $110 million total war chest on the midterms. The investment targeted competitive swing districts where industry-aligned incumbents who opposed federal AI regulation were vulnerable. Several of those incumbents lost. The incoming freshman class now includes a bloc of members who have committed to supporting federal AI oversight legislation expected to reach the floor in 2027.

Background: Since the EU AI Act took effect in 2024, the United States has remained the only major AI-producing nation without binding federal AI rules. Repeated attempts at federal legislation — including the AI Accountability Act and the Frontier AI Safety Act — stalled in committee due to opposition from tech industry lobbyists and members who argued federal rules would disadvantage U.S. companies relative to China. The 2026 midterms became, in part, a referendum on that position.

Public First's approach was to treat AI safety as a voter-facing issue rather than a technocratic one — running ads linking unchecked AI to job displacement, deepfake fraud, and financial system risk in districts where those concerns polled strongly.

The specifics:

• Total AI safety advocacy spend: $110 million across multiple organizations in the 2026 cycle
• Public First alone: More than $63 million, making it one of the largest single-issue PAC spends in this cycle
• Dario Amodei's role: Amodei, who co-founded Anthropic after leaving OpenAI, has publicly argued that the AI safety movement needs political power, not just research papers
• Congressional impact: AI safety-aligned candidates won in several key swing districts; industry-aligned incumbents lost in multiple races
• Legislative timeline: First federal AI oversight bill expected to reach the floor in early 2027
• Exact seat counts were not fully tallied as of publication time

What this means: Money in politics does not guarantee legislative outcomes, but it does change negotiating leverage. The $110 million campaign did two things simultaneously: it elected new members who will vote for oversight bills, and it sent a message to remaining industry-aligned members that opposing AI regulation is now a political liability in competitive districts.

For AI companies, the midterm results complicate the lobbying calculus. The industry's preferred approach — voluntary commitments and self-regulation — just became harder to defend when the electorate has signaled it wants enforceable rules. Companies like Anthropic, which backed the campaign, are betting that good-faith engagement with federal regulation is better for the industry long-term than a Wild West that invites a harsher backlash. Companies that have lobbied against regulation may now need to adjust their positions.

The Anthropic angle: It is worth noting the tension here. Anthropic is itself a leading AI developer deploying increasingly powerful models. Amodei's political spending reflects a genuine belief that AI poses risks worth regulating — but it also positions Anthropic favorably in a regulatory environment it helped shape. Competitors with less compliance infrastructure are more disadvantaged by new federal rules than a company that has built safety processes into its core operations from the start.

What to watch: The first real test will come in January 2027 when the new Congress convenes and committee assignments are made. Watch which members land on the House Energy and Commerce Committee and Senate Commerce Committee — the two bodies that will write AI legislation. Also watch whether the White House, which has taken a deregulatory stance on AI, shifts its posture in response to the midterm results or digs in for a veto fight.