OpenAI Delays IPO Until Models Reach Higher Safety Standards

OpenAI will hold off on an initial public offering until the company can make better promises about model safety. CEO Sam Altman provided no firm timeline for when an IPO might happen during a Q&A with reporters after his DevDay keynote on Tuesday.

Altman noted that OpenAI intends to continue with AI progress. He pointed to a recent surge in model capabilities and said the company must be able to make confident safety claims as more capability lies ahead.

Safety concerns delay public offering plans

Altman explained that going public could create the potential to disappoint Wall Street supporters in the names of safety. He stated that shifting to very capable models and new safety requirements makes an immediate public offering ill-advised.

OpenAI will prioritize safety to scale to the next stage of AI without people debating potential risks. Altman stated he does not want to put additional pressure on the company during this significant change.

Industry rivals move forward with public markets

Altman commented that waiting too long for an IPO would be bad for the world. He acknowledged that it is great when companies are public.

Rival company Anthropic officially filed to go public in June, with an IPO likely in November reportedly after the US midterm elections. Elon Musk’s SpaceX, which owns his AI company xAI, went public in June and became the biggest IPO in history.

Altman refrained from calling safety pacing an outright slowdown. He defined pacing as pushing safety and alignment ahead of capabilities.