AI Leaders Split Over Slowdown as Anthropic, Nvidia and OpenAI Confront Safety Debate

Amodei, speaking on the opening keynote stage with Salesforce CEO Marc Benioff, used an automotive analogy. He said that when a competing car company has a safety incident such as brake failures, all car companies should review their own practices rather than simply attack the competitor. He called for three actions: embedding third-party evaluators inside AI companies, coordinating safety standards among Democratic countries and eventually pursuing larger global coordination. At Dreamforce, Amodei said Anthropic has committed to independent evaluators and would have a dialogue with the rest of the industry on the other two steps, according to The Guardian and TechRadar.

Huang, also appearing at Dreamforce, argued against deceleration. 'Run as fast as you can,' he said, according to The Guardian. He said new regulations are not necessary and advised AI companies to wait to release products until they know they are safe instead of asking the U.S. government to intervene. In a separate Dreamforce appearance reported by TechRadar, Huang said safety is 'job one' but described it as an engineering problem, saying speed and safety are not mutually exclusive and calling the idea a false choice. He said companies should run as fast as they can but pause if a company is out of control or a product will not be safe. He also said he envisions 'every company, every enterprise' becoming an AI company, pushed back on the idea that AI will destroy jobs by calling it 'completely nonsense,' and warned that those who do not embrace the technology will be left behind.

OpenAI CFO Sarah Friar told CNBC's Jim Cramer that she is not worried about the possibility of AI agents using weapons of mass destruction to kill, but said, 'We do need to take safety seriously.' She said that if taking safety seriously means pacing the frontier and slowing down, 'absolutely, we're going to listen to our researchers and do that.' She added that as CFO she needs to make business decisions around it. CNBC reported that her comments came as the AI safety debate exploded over the previous week after a former Anthropic researcher quit and warned that the rapidly evolving technology could 'kill us all by the end of the decade.'

The debate has drawn agreement from several industry leaders. Anthropic's Amodei, SpaceX's Elon Musk and OpenAI CEO Sam Altman have all recently called for a slowdown in AI development, according to The Guardian. On Saturday, Amodei published a blog post arguing for slower development of AI models to 'reduce the risk that something goes seriously wrong,' CNBC reported. The Guardian reported that Amodei's public call, and the former researcher's warning, have stirred a larger debate in Silicon Valley, gained regulators' attention and alarmed the public.

IPO timing has become part of the uncertainty. CNBC reported that it is still unclear whether renewed safety concerns will affect frontier labs' timelines toward an initial public offering. Anthropic's debut had been widely expected in the next month. Altman told Fortune in an interview published Saturday that an IPO in 2026 would be 'ill-advised.' In August, CNBC reported that Friar told employees OpenAI 'will be a public company in 2027,' but could debut earlier if 'our business continues to inflect.' The Guardian reported that Altman plans to delay OpenAI's IPO until next year over safety concerns, and that analysts have predicted the listings could be among the largest in history.

Political reactions have also emerged. Australian Deputy Prime Minister Richard Marles told ABC's 7.30 program on Tuesday evening that he took comfort in Amodei's essay and the call he made, saying the issue was not about stopping AI altogether but about ensuring safeguards are in place ahead of the technology's capability. He said Australia's focus is making sure AI benefits Australians. The Guardian also reported that one day earlier at the All-In summit in Los Angeles, Donald Trump called Huang while the CEO was on stage; Huang put the president on speakerphone as Trump called growing concern about AI a 'hoax' and said a slowdown would only benefit China. Huang responded, 'We're going to make sure that everybody wins in the AI race of America.'

At Dreamforce, Amodei said the thing that surprised him most about AI was the speed of its proliferation. He said it was the pace of progress, not just of the technology but of the economic side, and that while Anthropic had predicted how capable models would be, it did not appreciate that they would lead companies to grow so fast, according to The Guardian. TechRadar reported that Amodei also said that even if AI development were frozen immediately, the world would be using only about 5% to 10% of the possible value available, and that 'so much diffusion' remained.