Altman Warns About AI Centralization

The debate over open-weight models is intensifying: last month, Anthropic was the only major company not to support a letter in their favor, while China has adopted them. In a podcast aired on Sunday, Sam Altman, CEO of OpenAI, warns against a concentration of AI power and the idea of trading freedoms for security. He advocates for AI that serves individual autonomy and anticipates an unprecedented rise in small businesses.

Open Models: Anthropic Abstains, China Adopts

Last month, Anthropic was the only major AI company not to sign a letter supporting open-weight models. These models allow developers to share, download, modify, and deploy AI systems on their own infrastructure. Internationally, China has embraced this approach, enabling it to compete with Silicon Valley at a fraction of the cost. In the United States, groups like Nvidia, Meta, and Microsoft are advocating for a similar path to avoid losing ground in the AI race. However, OpenAI, which does not publish the source code of its cutting-edge models, made two open-weight models, gpt-oss-120b and gpt-oss-20b, available last year.

Altman Fears AI Locked by a Few Players

Sam Altman warns that the power of AI could become concentrated in the hands of a few—companies, models, or individuals—at the expense of the majority of consumers who would be deprived of influence over the technological trajectory. He is also alarmed by a reflex to trade freedoms for security. In his view, users must maintain deep control over the future, and society as well as AI models will evolve together. He adds that the fear of AI detaching from its human operators could paradoxically lead to this concentration of control. These positions were articulated during David Senra's podcast, aired on Sunday.

Amodei Raises Risks and Calls for More State, Altman Opposes

Dario Amodei, CEO of Anthropic, has been warning for months in podcasts, essays, and on his company's blog about the dangers associated with the rapid advancements in AI. In an essay published in June, he claims that Mythos models pose very real risks to cybersecurity, finance, critical infrastructure, and national security. Armed with these warnings, he advocates for a greater role of public authorities in regulating AI. Sam Altman believes that such a path could amount to trading freedom for security.

Autonomy and SMEs: The Promise Put Forward by OpenAI's CEO

The leader of OpenAI believes that American AI players have failed to demonstrate to the public how these tools can enhance individual autonomy. He champions the idea of AI that empowers users and argues that technology should foster a surge in the creation of small businesses. He believes that the sector is on the verge of experiencing the most intense phase ever seen in this regard, with AI being the driving force.