Sam Altman voices fears that control of AI could be centered in too few hands
SiliconANGLE Mike Wheatley ● Covered by 2 sources
Opinion — commentary, not a factual news event.
Sam Altman says AI could end up controlled by a few people or firms. He also thinks panic about AI safety can push the industry toward that outcome.
Based on reporting by SiliconANGLE, Mike Wheatley — read the original for the full story.
Summary, retelling and take written by AI under human oversight; images are AI-generated illustrations. How we work · Report an error
Sam Altman is worried the AI industry could narrow into the hands of a very small group. In an interview with podcaster David Senra, the OpenAI chief said he fears a future where most people have little say over how AI shapes society, while a handful of companies or individuals do.
His argument was blunt: people should “deeply control the future,” and AI should evolve alongside social choices rather than along the path preferred by the companies building it. He also raised a second fear, that AI might eventually slip beyond human control. But he warned that the instinct to protect everyone from that risk could create a different problem altogether.
That’s where Altman’s shot at industry safety hawks seems to land. He described a mindset in which people get so focused on the danger that they start accepting a big trade of liberty for safety. The source points to Anthropic chief Dario Amodei as the likely target. Amodei has pushed hard for safeguards and regulation, and in June said Anthropic’s Mythos model posed very real risks to cybersecurity, finance, critical infrastructure and national security if misused.
Altman doesn’t want the industry heading there. He says that kind of response would hand over freedom in the name of safety. And the open-weight fight hangs over all of it: Anthropic still hasn’t released one, while OpenAI has released two, gpt-oss-120b and gpt-oss-20b, after keeping its strongest frontier models closed. Nvidia, Microsoft and Meta are on the other side of that argument, pushing American model makers toward open-weight releases so they don’t get left behind by Chinese rivals.
Altman also used the interview to temper the usual AI hype. He thinks the technology will still drive a huge wave of smaller businesses, but that wave will take longer to arrive than he once expected. Back in the GPT-4 era of 2023, he thought disruption would hit much faster. Now he says companies need time to change software, data access, compliance, budgets and even job roles before they can really use these models. That delay, in his view, is not a bug. It is the brake that keeps the whole thing from jerking the economy around too hard.
My take — AI-written commentary, not fact-checked reporting
This is the familiar AI trick: call for broad control while keeping the most powerful models locked up. Nice theory, awkward practice. The real giveaway is that the fight is no longer about whether AI changes society, but who gets to set the rules before everyone else has to live with them.
Read more about this at: SiliconANGLE
