OpenAI‘s chief, Sam Altman, recently admitted that his AI company might need a new open-source strategy. During an AMA on Reddit, Inc., he expressed concerns that OpenAI has been “on the wrong side of history” in terms of transparency with its technology.
Altman is personally in favor of open-sourcing their research, and this is a hot topic at OpenAI’s headquarters in San Francisco. However, it’s not currently their highest priority, and not everyone in the company agrees with him.
This discussion comes in the wake of Chinese AI newcomer DeepSeek AI making waves with its R1 chatbot, which claims to be an open-source project with low cost and high performance. Open source involves making the source code available to the public, which contrasts with private companies’ interests in revenue and intellectual property protection.
Meta, DeepSeek, and France’s Mistral AI differentiate themselves by allowing developers free access to their tools’ inner workings. When asked if DeepSeek influenced OpenAI’s future plans, Altman acknowledged that DeepSeek is a strong model, but OpenAI aims to produce even better ones, albeit with a smaller lead than before.