OpenAI employees donate $245,000 to counter their president's AI regulation stance

OpenAI Workers Challenge Leadership on AI Regulation

A remarkable split has emerged within OpenAI as current and former employees donated more than $245,000 to Guardrails Alliance, an AI safety super PAC explicitly created to counter Leading the Future, a pro-industry political group backed by the company's own president 1

1

5

5

. Eight OpenAI workers contributed to the effort, with research engineer Juan Felipe Cerón Uribe providing the largest single donation of $200,000 1

1

. The internal dissent reveals growing tensions inside the company over its approach to AI policy and the role of OpenAI president and cofounder Greg Brockman, who along with his wife Anna committed $25 million to Leading the Future 1

1

5

5

.

Cerón Uribe, who spent four years developing strategies to mitigate societal harms caused by AI at OpenAI, expressed concern that years of research "will have gone to waste if it doesn't translate to guardrails that hold private companies accountable for the responsible development of AI" 1

1

. He directly criticized Brockman for funding Leading the Future "to keep AI unregulated" 5

5

. Gabriel Wu, an OpenAI safety researcher who donated $5,000, cited concerns about "massive amounts of political money being spent to ensure AI remains unregulated" 1

1

5

5

.

Battle for Control Over AI Policy Takes Shape

The clash between Guardrails Alliance and Leading the Future represents a broader battle over how AI regulation will take form in the United States. AI companies spending on elections has reached unprecedented levels, with the two biggest artificial intelligence political action committees dropping at least $44 million into 40 House and Senate candidates by the end of June 2

2

. Together, these groups have raised more than $200 million to influence the 2026 midterm elections 2

2

.

Leading the Future launched last summer with backing from Brockman, venture capital firm Andreessen Horowitz, AI firm Perplexity, and venture capitalists Joe Lonsdale and Ron Conway 1

1

5

5

. The super PAC raised $125 million by the end of 2025 and spent more than $24 million on primary races through June 2

2

. Its stated goal is to "oppose policies that stifle innovation" and support candidates who back a national AI regulatory framework 1

1

3

3

.

Guardrails Alliance launched last month with $5 million in initial funding, aiming to raise $15 million this election cycle 1

1

. The group supports candidates pushing for stricter AI regulations and positions itself as representing parents, unions, and tech workers against "Trump-aligned AI billionaires" trying to influence elections 5

5

. Cofounder Shaunna Thomas argues the group doesn't need to match opponents dollar-for-dollar because "when you expose what the AI PACs are doing, the people reject it" 1

1

.

Industry Push for Federal Preemption Drives Political Spending

The central demand driving AI companies spending on elections is remarkably consistent: a single national framework for AI that preempts state-by-state regulation 3

3

. The industry argues that 50 different regulatory regimes would slow development and hand an advantage to China 3

3

. This push intensified after the Senate stripped a federal preemption provision by 99 votes to 1 before a bill was signed 3

3

.

Having lost that fight in Congress, the industry shifted to the ballot box. States introduced well over a thousand AI bills in 2025 alone, creating exactly the patchwork scenario the industry seeks to prevent 3

3

. The strategy mirrors successful efforts by crypto-backed PAC Fairshake, which spent $200 million in the 2024 elections and helped pass major stablecoin legislation 2

2

.

The battle played out dramatically in New York's 12th Congressional District race to replace retiring Rep. Jerry Nadler. Leading the Future spent more than $8 million opposing Alex Bores, who authored New York's landmark AI safety law 1

1

4

4

. AI safety groups affiliated with Public First Action poured over $20 million backing Bores, who ultimately lost to Micah Lasher in the Democratic primary 4

4

.

Anthropic Charts Different Course on Responsible AI Governance

Not every AI firm approaches regulation the same way. Anthropic CEO Dario Amodei gave $1 million to Public First, a super PAC backing candidates who support stronger guardrails on AI 4

4

. Several other Anthropic employees contributed a combined $2.15 million over the last quarter 4

4

. The company also gave $20 million to Public First Action, though that amount is restricted to educating the public on AI policy rather than political campaigning 2

2

3

3

.

Public First Action raised $80 million through the end of June and has spent $20 million on races so far 2

2

. The group has received donations from employees of OpenAI, Google, DeepMind and X 2

2

. Notably, John O'Farrell, a former general partner at Andreessen Horowitz—a major donor to Leading the Future—contributed to Guardrails Alliance, expressing sadness that "some of my former partners and friends" are using "hundreds of millions of dollars to be wielded to scare off any meaningful debate on the governing of AI" 5

5

.

The campaign finance battle will test whether voters accept the industry's frame that AI rules should be national, light, and set with companies in the room 3

3

. Watchdog groups have branded the wave "dark money" and warned of corporations trying to buy favorable regulation before the technology is fully understood 3

3

. As concerns about powerful AI models and unchecked AI development intensify, the political spending aims to influence AI regulation before the first national legislation takes final form 2

2

. For OpenAI employees funding the opposition, the stakes are clear: ensuring years of safety research translates into meaningful accountability rather than innovation and governance determined solely by industry interests.