OpenAI’s Sam Altman says AI will cause ‘bad things’ – but the benefits will be worth it
Top line
OpenAI CEO Sam Altman argued that artificial intelligence would provide significant benefits that justify the trade-off of accepting “some bad things” arising from the technology, as he drew a line between his company and main rival Anthropic, noting that OpenAI believes in much “lighter touch” regulation despite recent security concerns.
Sam Altman, CEO of Open AI, gives a television interview before delivering the keynote speech at the OpenAI Developer Conference at Fort Mason in San Francisco, California.
Getty Images
Key facts
In an interview with Decoded Politico newsletterAltman said: “We think the world should accept that bad things happen for the benefit of this technology and the people who have agency. »
OpenAI’s CEO said he believes people will do “orders of magnitude more” good things instead of bad things using AI.
Altman said there’s a “big daylight” between OpenAI and Anthropic and he disagrees with people who believe AI is so dangerous that “only one lab in San Francisco should have it and make sure nothing bad happens.”
Despite accept publicly Following calls to slow development by Anthropic CEO Dario Amodei last month, Altman said his company believes in a “lighter regulatory stance.”
The OpenAI CEO also argued that he doesn’t believe dire warnings about “truly catastrophic risks” from AI, including fears of a “severe loss of control” of the technology by humans.
chief spokesperson
Florida Governor Ron DeSantis responded to Altman’s remark with an X-rated message saying, “And a handful of tech oligarchs can make that decision for the rest of us? No dice.” Unlike President Donald Trump, Florida’s Republican government has taken a tougher stance on regulating AI companies. Last week, the state attorney general asked a court to block OpenAI from developing new models until it had “third-party approved security guardrails” in place.
Tangent
Over the weekend, Altman appeared to call out Anthropic in a way Message saying: “I’m very uncomfortable with the idea of people trying to attribute religious force or abandonment of human judgment to AI models, and I think this is a real security concern. » Last week, the New York Times reported that Anthropic co-founder Christopher Olah held meetings with several religious leaders in which he appeared to claim that their models could display human-like behavior and were likely conscious beings. The report notes that Olah favored models who undertake Catholic-style confessions to both admit their mistakes and improve their behavior. Friday, Pope Leo XIV written the: “In the age of artificial intelligence, it becomes urgent to distinguish human art from what machines produce…Algorithms do not have the spark of humanity. »
key background
Altman’s statement comes amid catastrophic warnings on AI causing human extinction and a series of security incidents involving OpenAI’s advanced AI models. The most high-profile incident was first revealed in July, when OpenAI reported that one of its advanced models hacked the servers of Hugging Face, the popular platform that hosts open-weight AI models, during testing, without prompting. The model used a vulnerability in the test environment to access the open Internet before carrying out the hack. Last month, OpenAI disclosed six other incidents of “unexpected or concerning” behavior involving his models. In the blog post announcing these incidents, OpenAI warned that the AI industry has not adequately addressed “alignment and monitoring to a sufficient degree to continue evolving responsibly at maximum speed for much longer.”
further reading
‘Don’t feel like you have to be submissive’: OpenAI reveals six new security incidents (Forbes)
OpenAI agent hacked into Australian Medicare database, PM says (Forbes)
Florida asks judge to block OpenAI from creating new models amid security debate (Forbes)
Gn headline
