OpenAI CEO says the company is being held back by lack of compute capacity

OpenAI chief executive Sam Altman said on Thursday that the company was restrained by a lack of compute capacity in shipping products as often as it would like.

“All of these models have gotten quite complex and we can’t ship as many things in parallel as we’d like to,” Altman wrote in response to a question about why GPT-5 was taking so long. “We also face a lot of limitations and hard decisions about we allocated our compute towards many great ideas.”

The Microsoft (NASDAQ:MSFT)-backed startup has reportedly struggled to secure enough computing power to build and run its AI models. It is reportedly working with Broadcom (AVGO) and TSMC (TSM) to design its own chips.

Altman reiterated that AI startup won’t be releasing its next big AI model this year. OpenAI has “some very good releases coming later this year” but “nothing that we are going to call GPT-5.”

Kevin Weil, OpenAI’s chief product officer, said the company needed “to perfect the model, need to get safety/impersonation/other things right, and need to scale compute!” before launching the next major upgrade for its video model Sora.

Altman added that the startup’s image model DALL-E had no planned release yet, adding that the next model will be “be worth the wait.”

Last week, the company had denied a report by the Verge stating that it was planning to release an AI model code-named Orion this year.

The company’s “Ask Me Anything” session on Reddit (NYSE:RDDT) followed the launch of a new set of search features to ChatGPT, heating up its competition with Alphabet’s (GOOG) (GOOGL) Google.

OpenAI also addressed its string of recent high-profile executive departures, with ​​Srinivas Narayanan, VP Engineering, stating, “While we are sad to not have some of the people we had worked with closely, we have an incredibly talented team and many new amazing people who have joined us recently as well.”