Not everyone is convinced that Big AI's proposed development slowdown is really about safety

Details

Anthropic CEO Dario Amodei formally called on the AI industry and governments to coordinate a slowdown of frontier AI development, seeking antitrust exemptions to allow major labs to set shared safety standards. OpenAI's Sam Altman and Elon Musk publicly agreed with the proposal. Critics are pushing back hard. Cohere co-founder and CEO Aidan Gomez published a detailed rebuttal, arguing the proposal would erect massive barriers to entry — requiring huge compute, monitoring infrastructure, safety organizations, and embedded evaluation teams — effectively locking out any challenger. The few 'independent' evaluation bodies mentioned are funded by the same companies they would audit. Gomez calls the safety warnings a smokescreen while real harms like voice-cloning fraud go unaddressed. From the open-source side, Hugging Face engineer Niels Rogge dismissed the statements as 'the most bizarre nonsense' he has read, arguing Amodei is primarily worried about open-weight competitors like DeepSeek and Z.ai undermining the bet-on-scale strategy. Politically, President Trump dismissed AI takeover warnings as a hoax and the White House opposes the initiative. Democrats including Barack Obama and Kamala Harris have voiced support for a slowdown and more transparency. The divide signals a major regulatory battle ahead for frontier AI governance.