Sam Altman Admits ChatGPT Has Weaknesses, Says Gemini 3 Helped Spot Them – Mashable India
OpenAI CEO Sam Altman has admitted that intensifying competition in the AI industry is putting sustained pressure on the company to move faster and constantly reassess its priorities. Speaking on the Big Technology Podcast, Altman revealed that OpenAI has repeatedly entered “code red” mode in response to rival AI models narrowing the gap with ChatGPT, a level of urgency he says is becoming the new normal as the AI race heats up.
Sam Altman reveals OpenAI goes into “code red” 1-2 times per year after competitive threats like Gemini and DeepSeek, treating them as “six or eight week things” – says ChatGPT lead will “increase not decrease over time” despite competition intensifying” pic.twitter.com/br98An5qjo — Victor (@victor_explore) December 19, 2025
Sam Altman said OpenAI has adopted an aggressive and highly responsive mindset whenever a potential competitive threat emerges. Speaking on the podcast, he explained that being “paranoid” and acting quickly is now embedded in how the company operates, noting that such emergency “code red” phases could occur once or twice a year as OpenAI works to stay ahead in an increasingly fast-moving AI market.
ALSO SEE: Croma Slashes iPhone 15 Price to ₹36,490 During Cromtastic December Sale
One major code red moment earlier this year was triggered by the rapid rise of China-based DeepSeek. The company drew global attention in January after claiming its AI model could rival leading systems, including OpenAI’s o1 model, while operating at significantly lower costs. Altman said this unexpected development forced OpenAI to reassess its strategy and respond with urgency.
Sam Altman just confirmed OpenAI triggers “code red” 1-2 times a year.
This happens whenever competitors like Gemini 3 or DeepSeek land a hit.
He treats these as “6 to 8 week” sprints to fix product weaknesses.
His mindset:
➧ Paranoia is necessary to win
➧ Competitive… pic.twitter.com/CGfRRMRVEU — Abu (@abuchanlife) December 20, 2025
More recently, OpenAI entered another code red phase following the rollout of Google’s Gemini 3 AI chatbot. Although Gemini 3 was launched in November, its growing traction in recent weeks raised internal concerns. Altman reportedly told employees that ChatGPT would become OpenAI’s top priority, even if it meant slowing or delaying other product initiatives.
Altman later clarified that Gemini 3 did not disrupt the market as much as initially feared, but it did highlight gaps in OpenAI’s broader product strategy. He said these weaknesses are already being addressed, pointing to recent updates including a more capable model for professional tasks and a new image-generation system. While such emergency phases typically last six to eight weeks, Altman noted that code red responses are common across the industry, recalling how Google declared a similar internal alert after ChatGPT’s public launch in 2022.
