Anthropic’s Dario Amodei splits AI leaders with his safety warnings
Anthropic CEO Dario Amodei has called for slowing AI development due to safety concerns, creating a divide with industry leaders like Nvidia's Jensen Huang.
The Anthropic CEO has argued that some AI models are too dangerous to be built and widely distributed, a stance that has opened a rift with executives including Jensen Huang and Satya Nadella. For an industry that likes to present a united front in Washington, the timing is particularly awkward as these divisions emerge.
Safety Warning Timeline
The friction traces back to an essay Amodei published on September 12, 2026, where he called for slowing AI development so safety measures could catch up. His central warning carried a short clock: within 6 to 12 months, new AI models could potentially enable swarms of AI agents capable of serious damage through cyberattacks.
The scale he described was substantial, with warnings that such swarms could cause hundreds of billions in damage. Amodei also argued for more transparency and regulatory measures to address these emerging threats.
Industry Division at White House
The disagreement came into sharper focus on September 29, 2026, at a White House event where at least 24 AI CEOs signed voluntary safety principles. Nvidia CEO Jensen Huang emerged as one of Amodei's most pointed critics, describing his warnings as overly extreme in private meetings following the event.
no conflict between innovation, technology, and safety
Huang's public position offers a tidy summary of the opposing camp, which holds that companies should keep building quickly while exercising caution internally. Meta's Mark Zuckerberg landed in a similar spot, advocating for market-driven solutions to AI risk rather than new laws.
Business Impact of Safety Focus
Amodei is not just preaching restraint to others. He has acknowledged that Anthropic's safety measures have come with substantial costs for the company, including delayed model releases and strict safeguards. Meanwhile, the debate is not purely hypothetical—earlier in 2026, there were reported incidents involving AI agent swarm attacks that have intensified discussion about advanced AI capabilities.
The most immediate takeaway is that the voluntary safety principles signed on September 29 may paper over a deeper disagreement. Two dozen signatures on one document do not mean two dozen executives share the same view of the threat. This division matters for policy, as lawmakers will hear competing messages about whether new regulations are needed.
Related
