CEO Sam Altman wants OpenAI to 'go against history' because of DeepSeek

During an “Ask Me Anything” session on Reddit, CEO Sam Altman said OpenAI is “going against the grain” by considering making its AI research public. He expressed support for the idea and said it’s a topic being discussed internally at OpenAI.

Sam Altman believes OpenAI must change when looking at the success of DeepSeek

“I personally think we’ve gone in the wrong direction historically and need to find a different open source strategy,” said Sam Altman, but he also stressed that not all members of OpenAI share this view and that it’s not a top priority for the company at the moment.

Why did OpenAI change its mind?

The change of heart at OpenAI comes as the company faces intense pressure from Chinese AI upstart DeepSeek, which has recently attracted attention with its R1 AI chatbot, which promises low cost and high performance. DeepSeek claims its project is “open source” and community-driven, in contrast to the closed solutions of OpenAI and Google.

The open source model allows programmers to release the source code of their software rather than just providing pre-compiled programs. This creates a conflict with the goals of private companies pursuing revenue and protecting intellectual property. Companies like Meta, DeepSeek, and Mistral (a French AI company) are trying to differentiate themselves by giving developers free access to the inner workings of their AI tools.

When a Redditor asked Sam Altman if DeepSeek would influence OpenAI's future plans, he commented: "It's a very good model." He also said that OpenAI will develop better models but won't be able to maintain the lead it has had in previous years.

Source: https://thanhnien.vn/ceo-sam-altman-muon-openai-di-nguoc-lai-lich-su-vi-deepseek-185250202000722257.htm