Dario Amodei, CEO of Anthropic, has called for slowing down the development of powerful AI models, opening a new front in the technology competition between the US and China and highlighting the challenges of collaboration between the two leading AI powers in the realm of safety.
Amodei's Proposal to Governments and Companies
In a recently published article, Amodei argues that rapid advancements in AI, particularly systems that can aid in the development of more advanced versions of themselves, may outpace researchers' ability to understand and control them. He stated, "We need to reduce the pace of improvement of AI model capabilities. Progress still seems fast, and we should use the time we gain wisely."
Read more: An Unmatched Experience with the iPhone Duo: Apple's Major Change in Phone History
His proposal includes independent assessments within leading AI companies, coordination among companies and governments in democratic countries, and ultimately international agreements to regulate the pace of AI development.
Conflicting Reactions from US and Chinese Officials
Leaders of major AI companies such as Sam Altman from OpenAI and Demis Hassabis from Google DeepMind supported some aspects of Amodei's proposal, but the government of Donald Trump, the former US President, viewed the idea as a geopolitical problem and strongly opposed it.
Trump emphasized, "We are ahead of China in AI. We are the most advanced country in the world, and I want to maintain that status because whoever wins in AI will win." He also criticized why industry leaders should seek regulations that might limit their own companies.
China has also responded to this proposal with skepticism, with the spokesperson for the Ministry of Foreign Affairs, Guo Jiakun, stating that "fear and ruthless competition only undermine efforts for global AI governance and will benefit no one."
This discussion comes as reports indicate that the US is still leading in AI, but Chinese companies are closing the technological gaps. Chinese developers like DeepSeek, Moonshot AI, and Alibaba have produced more competitive models, although researchers in China generally believe that American labs are ahead in advanced AI capabilities.
Read more: David Kruger: AI Development Should Be Halted · Greg Stanton Warns: AI Could Endanger All Humanity
