IT之家 November 6 news: OpenAI CEO Sam Altman (IT之家 note: Sam Altman) recently said while appearing on the Raj Shamani podcast that artificial intelligence could spiral out of control in two ways: one is humans gradually handing control over to autonomous AI, and the other is a monopoly on AI technology by a single person or entity.
Regarding the first scenario, Altman warned: "AI could gradually slip out of control; we may hand control over, intentionally or unintentionally, to systems far more capable than humans, and ultimately be unable to take it back."
The risk of the second scenario lies in monopoly by giants. Altman explained: "The other scenario is that some person, company, or country becomes too powerful relative to everyone else. Having multiple AI labs pursuing different approaches compete against each other is far safer than everyone depending on a single company."
This concern about excessive centralization explains why well-known figures such as Musk and former OpenAI executives have chosen to leave and found competing companies. Altman pointed out: "A thought that occurs in many people's minds is that only they can be trusted to develop this technology, because the stakes are simply too high."
That said, Altman still firmly believes AI will bring humanity a better future. He specifically mentioned basic scientific research, arguing that next-generation AI models can enable science to keep advancing.