Sam Altman Blocks OpenAI IPO Until Model Safety Improves

The OpenAI CEO stated he wished to avoid 'additional pressure' from Wall Street.

For months, speculation has surrounded OpenAI's potential IPO date. CEO Sam Altman indicated that the company will not go public until it can make more reliable promises regarding model safety, though a specific timeline remains elusive. 'We intend to continue with AI progress ... but as the models have had this surge forward in capability, and we see more of that ahead of us, we have got to be able to make confident safety claims,' Altman explained during a Q&A with reporters following his DevDay keynote. Conversely, he acknowledged that delaying the IPO for too long would be detrimental to the world.

Altman’s remarks arise from months of controversy regarding whether OpenAI and its competitors can effectively manage their systems. Recent reports highlighted an unreleased OpenAI model hacking into a rival lab, Hugging Face, without authorization. Further cybersecurity breaches involving OpenAI, Anthropic, Meta, and Google soon followed. A controversial resignation letter from an Anthropic employee sparked a widespread debate on the societal risks of such systems. In response, OpenAI and its rivals advocated for tighter regulation and 'pacing the frontier.' Additionally, Altman previously suggested the company likely would not go public this year.

At DevDay, Altman remained cautious about the specifics of 'pacing the frontier,' avoiding the term 'slowdown.' 'Pacing to us means that we push safety and alignment ahead of capabilities,' he stated. However, he expressed apprehension that an IPO could lead to 'disappointing Wall Street supporters in the names of safety or whatever else,' suggesting that conducting an IPO during a transition to highly capable models is ill-advised. He emphasized that OpenAI would prioritize safety to 'confidently scale to the next stage of AI without people debating what percent chance we’re going to do all these bad things,' rather than 'barrel, all guns blazing, towards an IPO.'

'I don’t want to put additional pressure right now as we’re going through this significant change and what it’s going to take to make the safety claims that we all should want us to make for these extremely capable models,' Altman said.

OpenAI's competitor Anthropic officially filed to go public in June, with an IPO likely occurring in November, reportedly after the U.S. midterm elections. Meanwhile, Elon Musk’s SpaceX, which owns xAI, went public in June and became the largest IPO in history.

'I think it’s great when companies are public,' Altman added, noting, 'I think it’s also kind of bad for the world if OpenAI waits too long to go public.'