- Published on
If you’re really worried about AI going out of control, stop it yourselves—don’t seek regulatory protection under the...
- Authors
- Name
- DeepHorus
September 14, 2026, according to cnBeta.COM, American tech investor and Chairman of the President's Council of Advisors on Science and Technology David Sacks publicly criticized the increasingly intense argument for slowing AI down. He believes that if OpenAI and Anthropic are truly worried that the current pace of AI development is dangerous, they should slow down their own R&D, rather than making industry coordination, antitrust exemptions, or government approval mechanisms additional conditions. These remarks further heated up the debate over whether frontier artificial intelligence should proactively slow down.
The background to this debate is Anthropic CEO Dario Amodei's public call for the entire AI industry to slow the pace of improving frontier model capabilities. OpenAI CEO Sam Altman and xAI CEO Elon Musk subsequently expressed support. In a short period, several of the world's most important AI companies rarely formed a considerable degree of consensus on the issue that there 'should be appropriate slowing.'
Amodei previously published an article warning that frontier AI capabilities are rapidly approaching a stage that could create major safety risks. He believes that if models continue to develop at the current pace while safety mechanisms cannot keep up in tandem, within the next 6 to 12 months, swarms of AI agents might even gain the ability to control the internet on a large scale and be used to launch cyberattacks or cause other serious harm. This judgment became the core basis for his call to slow down.
Amodei proposed a plan called 'slowing down frontier AI.' It includes allowing independent third-party safety assessment institutions to go deep inside AI companies and continuously inspect model safety with permissions close to those of employees; having major AI laboratories coordinate safety standards to avoid continuously lowering safety thresholds due to competitive pressure; and at the same time promoting closer cooperation between the United States and its allies, and ultimately discussing AI safety issues with other countries including China. The plan places industry coordination and regulatory arrangements in an important position.
But Sacks believes there is an obvious logical problem in this. In his view, if Amodei and Altman truly believe that next-generation AI models pose a risk serious enough to require slowing down, the two companies do not need to wait for government approval, nor do they need to persuade competitors, much less a new legal framework. As the currently most advanced AI development companies, they can simply directly reduce their own R&D and deployment speed.
Sacks even suggested that if these companies truly believe that certain AI systems are already dangerous enough to potentially threaten humanity, then the 'simplest way' is to stop developing or releasing those systems. He believes that instead of asking the government to establish a new regulatory system, it would be better to let the companies that truly possess frontier AI R&D capabilities take action themselves. This statement directly challenges the mainstream narrative of trading regulation for safety.
What Sacks truly opposes is the part of Amodei's plan involving industry coordination and regulatory arrangements. He warned that if competitors such as OpenAI and Anthropic obtain antitrust exemptions through the government and then jointly decide at what speed the entire industry should develop AI, such an arrangement could actually evolve into a kind of 'cartel.' In Sacks's view, allowing a few of the leading AI companies to jointly set the pace of frontier AI development could produce results completely different from safety goals, and existing giants could also use such a mechanism to restrict competition.
Currently, Altman has expressed agreement with Amodei's view that the AI industry needs to slow the pace of frontier model development, and said the issue has become an important topic in recent internal discussions at OpenAI. Musk has also publicly expressed agreement. Disagreements over AI safety, regulation, and competition continue, and Sacks's criticism has made 'whether to slow down' and 'who decides to slow down' unavoidable questions for the industry.
