OpenAI will not go public in 2026. CEO Sam Altman told Fortune the timing would be 'ill-advised,' citing active safety concerns as the primary reason to stay private.
The 45-minute interview covers more than the IPO delay. Altman addressed the Hugging Face hacking incident, recursive self-improvement risks, and directly confirmed that an AI beyond human control is 'absolutely' possible. He said OpenAI would pause training if necessary, framing it plainly: 'there are risks we should not be able to incur on behalf of humanity.'
The IPO headline is the hook, but the substance is Altman on paper drawing the line between acceptable and unacceptable AI development risk. Read the full Fortune interview to hear how he justifies continued scaling while simultaneously arguing for that same ceiling.
[READ ORIGINAL →]