OpenAI Will Not Go Public in 2026, According to Sam Altman

OpenAI does not plan to go public in 2026. Sam Altman explains that he feels no pressure and considers the timing inappropriate given security concerns, while detailing the precautions he says he is willing to take regarding model training.

Delayed Timeline and Inopportune IPO

Sam Altman believes that, given the current security landscape, it would be ill-advised to go public at this time. He states that OpenAI is not rushing toward an IPO and does not feel any pressure on this matter. He reminds that the company has long said it will do so when it is ready and clarifies that this will not happen in 2026. He adds that OpenAI still has a lot to accomplish.

Precautionary Thresholds Discussed for Training

Sam Altman considers it entirely possible to build an AI that could escape human control. He promises to take measures to prevent such a scenario, including suspending training if necessary. He further states that certain risks should not be taken in the name of humanity. These positions were articulated as he addressed the possibility of AI beyond human control.

A 45-Minute Interview Covering Multiple Topics

These statements were made during a 45-minute interview. Sam Altman notably discussed the Hugging Face hacking incident and recursive self-improvement. He also talked about the possibility of building AI beyond human control. Sam Altman is the CEO of OpenAI.