Sam Altman Discusses DeepSeek and its Impact on OpenAI

In a Reddit AMA (Ask Me Anything), OpenAI CEO Sam Altman admitted that China-based DeepSeek had undermined his company’s lead in AI. He also said that OpenAI was “on the wrong side of history” when it came to AI and open source, and that he wants to fix that.

“[DeepSeek] is a very good model,” Altman said in his first answer during the chat. “We will produce better models, but we will maintain less of a lead than we did in previous years.”

When asked why the recently released OpenAI o3-mini reasoning model–which mimics DeepSeek by showing itself “thinking” through how to answer questions–has a “knowledge cutoff of October 2023,” Altman said this is no longer an issue. “Now that we have enabled search, this matters much less,” he said. “In my own use, I never think about the knowledge cutoff anymore.” (OpenAI’s Hongyu Ren added that, while the company is working on updating its knowledge cutoffs, o3-mini can “browse the web.”)

When asked whether OpenAI would raise the price of the $20-per-month ChatGPT Plus subscription, Altman said he’d “like to reduce it over time.” (Another OpenAI representative simply answered “Nope” to the price increase question.)

Altman noted at one point that the greatest impact of AI will come through “accelerating the rate of scientific discovery, which I believe is what contributes most to improving the quality of life.” He said that the full o3 model will debut in “more than a few weeks, [but] less than a few months.” And that o3 Pro–this is the $200 per month model–will be “super worth it.”

But the money quote came when someone asked about OpenAI releasing model weights and publishing research? This is tied to OpenAI’s technology being proprietary, as opposed to an open source solution like DeepSeek that transparently explains how its technology works.

“We are discussing [that],” he answered. “I personally think we have been on the wrong side of history here and need to figure out a different open source strategy. Not everyone at OpenAI shares this view, and it’s also not our current highest priority.”

Normally, I would push that aside as a PR response to what’s perceived as a crisis. But with AI changing so rapidly, and OpenAI being among the most chaotic and aggressive companies in this space, it’s reasonable to believe that this will be a dramatically different company in the near future. We’ve already seen Microsoft add the OpenAI o1 model–which that company charges for–to Copilot for free. I bet there are other major changes coming too.