OpenAI CEO Admits Compute Shortages Delay Key Product Launches
OpenAI CEO Sam Altman reveals that limited computing power forces difficult trade-offs, slowing down the release of upcoming AI models and features. The hardware bottleneck directly impacts highly anticipated tools like Sora and DALL-E.
OpenAI CEO Sam Altman reveals in a recent Reddit AMA that a lack of compute capacity is a major factor delaying the company's product launches. As AI models grow increasingly complex, OpenAI faces difficult decisions regarding how to allocate its limited computing resources across numerous promising ideas. This infrastructure bottleneck directly prevents the company from shipping new products at the pace it desires.
The strained computing capacity forces OpenAI to postpone several highly anticipated features, including vision capabilities for ChatGPT's Advanced Voice Mode. Altman confirms that the next major release of the DALL-E image generator currently lacks a launch timeline. Furthermore, OpenAI's chief product officer notes that the Sora video generation tool remains held back by the need to scale compute, perfect the model, and resolve safety concerns.
To address these ongoing hardware limitations, OpenAI is reportedly collaborating with Broadcom to develop a custom AI chip that could arrive by 2026. Despite these setbacks, Altman emphasizes that the company's top priority remains improving its o1 series of reasoning models and their successors. He also hints that OpenAI is still considering allowing adult content in ChatGPT someday to treat users like adults.