Altman ties OpenAI IPO plans to safety claims
The OpenAI chief executive gave no timetable, saying public-market pressure could complicate safety priorities as the company develops more capable models.
OpenAI will not pursue an initial public offering until it can make confident safety claims about its increasingly capable models, chief executive Sam Altman said after his DevDay keynote. He gave no timetable or specific safety threshold.
Altman told reporters the company intended to continue advancing AI while putting safety and alignment ahead of capabilities. He said a shift towards more capable models would require new safety requirements.
An IPO could bring pressure to meet Wall Street expectations, Altman said, including the risk of disappointing investors by prioritising safety. He said he did not want to add that pressure during what he described as a significant change for the company.
Altman also cautioned that waiting too long to list could be bad for the world. His remarks follow his earlier comment this month that OpenAI was unlikely to go public this year.
The comments come amid public debate over whether AI developers can control increasingly capable systems. OpenAI and rivals have called for regulation and “pacing the frontier”; Altman said the phrase meant advancing safety and alignment ahead of capabilities, without describing it as an outright slowdown.
