Not everyone is convinced that Big AI's proposed development slowdown is really about safety
decoder ·
Major AI laboratories including OpenAI and Anthropic have proposed a coordinated slowdown in frontier AI development, citing safety concerns and requesting antitrust exemptions to implement the plan. However, this initiative has faced significant skepticism from industry leaders who argue it is a strategic move to entrench market dominance rather than a genuine safety effort. Aidan Gomez, CEO of Cohere, criticized the proposal as a cartel designed to create high barriers to entry that would disadvantage competitors and weaken open-weight AI models. Niels Rogge, an engineer at Hugging Face, dismissed the existential risk warnings as nonsense, suggesting the labs are primarily motivated by competitive fears regarding Chinese models and open-source alternatives. David Sacks, the White House AI czar, noted that while he supports a voluntary pause if risks are real, the companies may be seeking regulatory protection from product liability rather than acting altruistically. On the political front, President Donald Trump rejected AI takeover warnings as a hoax, aligning with his broader dismissal of such technological threats. Conversely, Democratic figures like Barack Obama and Kamala Harris have endorsed the call for a development slowdown and greater transparency. Meanwhile, China has reportedly rejected the proposed global agreement, viewing it as an attempt by the US to maintain its technological lead.
decoder