Key Notes
- Elon Musk, in a wide-ranging interview, elaborated on his earlier "Dario is right" post, saying he meant AI danger "is very significant right now" and that when many people at Anthropic and OpenAI say their models are dangerous, "we should believe them".
- Musk described a cyber-to-existential risk pathway: if AI models gain control over military systems or nuclear weapons, "that would be bad," and proposed that major AI competitors test each other's models before release, with anyone able to raise an alarm.
- This follows Anthropic researcher Jacob Coxon's resignation and CEO Dario Amodei's "We Must Pace the Frontier" essay, both publicly endorsed in some form by Musk, Sam Altman and Microsoft's Satya Nadella in the same week.
Elon Musk used a wide-ranging interview to elaborate on a brief social media post in which he wrote “Dario is right,” referring to Anthropic CEO Dario Amodei’s public call to slow the pace of frontier AI development.
“I probably should have said that Dario is right about… I did try to clarify in subsequent posts on X, but they get far less attention,” Musk said. “What I meant by him being right is that AI danger is very significant right now.”
Musk argued AI safety practices need to improve substantially, warning that failing to do so risks confronting “exponentially growing risk” tied to increasingly capable models.
He said this view was shaped by conversations beyond Amodei himself.
“I’ve heard this not just from Dario, but from a lot of other people at Anthropic,” Musk said. “When a lot of people from Anthropic and OpenAI tell you their models are very dangerous, I think we should believe them.”
Musk described what he sees as the pathway from cyber risk to existential risk, saying that if AI models gained control over military systems or nuclear weapons, “that would be bad.” He characterized navigating this moment as “some kind of crazy 4D chess game,” referencing a figure some in the industry cite of roughly a 10% probability of human extinction from advanced AI.
As a concrete safety proposal, Musk said the largest AI competitors should test each other’s models before public release, describing a system in which “everyone checks everyone’s models” and any participant who sees something concerning would raise an alarm.
Why This Fits a Broader Pattern
Musk’s remarks build on a public statement he made on X, responding directly to Amodei’s essay “We Must Pace the Frontier,” in which the Anthropic CEO argued the industry needs to slow capability gains to keep safety work from falling behind.
Amodei’s essay followed the resignation of Anthropic researcher Jacob Coxon, who publicly accused both Anthropic and OpenAI of racing toward self-improving superintelligence “without acting responsibly,” and later told CBS News he believes the risk resembles Hollywood’s darkest depictions of machine takeover.
OpenAI CEO Sam Altman also backed Amodei’s call for slower development, and Microsoft CEO Satya Nadella said the company “welcomes the research, focus, and deliberate pacing needed to get alignment right,” with Microsoft subsequently publishing a draft code of conduct for its own AI models.
Amodei has separately acknowledged the difficulty of his position, telling CBS News that slowing down unilaterally while competitors, including those in China, continue advancing is “the toughest dilemma” the industry faces, and pointed to Cold War-era nuclear arms agreements as a possible model for future AI governance between nations.
Musk’s own public position carries some tension worth noting: his company xAI, along with OpenAI, Amazon and Microsoft, continues to sell AI services to the US military, even as Musk has separately invoked “Terminator”-style scenarios about AI and nuclear escalation in other public statements, including in his ongoing lawsuit against OpenAI.
Disclaimer: AIstify is an independent media brand owned and operated by NuvexMedia LLC, publishing news, research, and insights on artificial intelligence, emerging technologies, automation, and related industries. NuvexMedia LLC invests in and collaborates with companies across the AI, technology, software, and digital innovation sectors. These relationships do not influence AIstify’s editorial coverage, and the publication maintains full editorial independence to provide accurate, timely, and objective information. © 2026 NuvexMedia LLC. All rights reserved. This content is for informational purposes only and should not be considered legal, tax, investment, financial, or other professional advice.
