Industry divisions deepen as AI leaders split on regulation and safety oversight

Anthropic CEO Dario Amodei outlined a three-step plan to slow AI development, including embedding third-party evaluators in labs, coordinating across the domestic industry, and pursuing international agreements, with potential government involvement. OpenAI CEO Sam Altman, Google DeepMind co-founder Demis Hassabis, and SpaceX CEO Elon Musk publicly expressed support for elements of the plan, framing it as both a PR move and a proactive step to address safety concerns amid high-profile hacking incidents.

However, the industry remains divided. Meta CEO Mark Zuckerberg, Nvidia CEO Jensen Huang, and Musk reportedly opposed proposals for an industry-funded independent regulator similar to the Financial Industry Regulatory Authority (FINRA), according to a Wall Street Journal report. Zuckerberg reiterated his stance on X, stating that each lab bears responsibility for training models safely without external constraints.

The Trump administration has publicly dismissed AI safety concerns as a 'hoax,' with President Trump asserting that existing regulatory and criminal powers are sufficient to control risks. Trump also claimed his administration had already prevented AI from causing harm, framing AI leadership as a geopolitical imperative.

OpenAI’s global affairs chief, Chris Lehane, countered this stance, arguing that while companies must take initial safety steps, government has a critical role in establishing mandatory national safety standards for frontier AI. Lehane emphasized bipartisan support in Congress for such measures, framing them as essential to balancing safety and innovation while maintaining U.S. leadership in AI.

An anonymous AI venture capitalist compared the current moment to historical technological transitions—such as electricity, cars, and airplanes—where industries eventually coalesced around safety standards after early unruliness. This perspective underscores the argument for coordinated regulation to ensure broader adoption and trust in AI technologies.

Anthropic declined to comment on Amodei’s proposal, but Amodei previously stated that the company has advocated for 'well-considered regulation of AI,' even when facing criticism of 'doomerism' or regulatory capture. Other companies, including Safe Superintelligence Inc., SpaceX, Google, and Thinking Machines Lab, did not respond to requests for comment.