Sam Altman Declares 'Code Red' at OpenAI Over Gemini

Sam Altman has declared an internal "code red" at OpenAI, according to reports that surfaced this week: a memo ordering a freeze on secondary ChatGPT projects — advertising, shopping features and the proactive assistant Pulse — so engineering and product teams can focus on improving the quality and speed of the company's flagship chatbot. The trigger is Gemini's momentum, Google's model, which according to the company's latest figures now reaches close to 650 million monthly users.

The symbolism here is hard to overstate. In the three years since ChatGPT launched in November 2022, OpenAI has been the unquestioned reference point in the public conversation around generative AI. Now it's OpenAI itself activating the emergency protocol that Google set in motion back in December 2022, when ChatGPT's sudden rise set off alarm bells inside a company that had led AI research for years without managing to turn it into a mass-market consumer product.

What a "Code Red" Means at OpenAI

The memo, whose contents leaked through internal sources, calls for a reshuffling of priorities: less energy on monetization and feature expansion — such as the in-app advertising OpenAI had been exploring for months, or the integrated shopping system with merchants the company unveiled a few months ago under the name Instant Checkout — and more focus on the product's core. Pulse, the feature that since September has offered paying subscribers proactive daily AI-generated summaries, is also being put on hold.

This is textbook crisis-management playbook: when a leadership position starts to feel threatened, side bets get trimmed and resources flow back into the product that carries the brand. That it's Altman signing off on this, rather than a mid-level product lead, signals that inside OpenAI the threat is being taken as real, not just a headline to shrug off.

Gemini Goes on the Offensive

The pressure comes barely two weeks after Google unveiled Gemini 3, the model that has revived direct comparisons with GPT-5, OpenAI's flagship system launched in August. Gemini 3's results across several public benchmarks and its warm reception among developers have reinforced the narrative that Google, after three years on the back foot, has closed the technical gap with OpenAI and is now competing on equal footing for the first time since ChatGPT's launch.

The 650 million monthly users Google has disclosed for the Gemini app puts the company, for the first time, in territory comparable to ChatGPT, which Altman had placed above 800 million weekly users back in October. These are different metrics — monthly versus weekly — so an exact comparison isn't possible, but the trend is what matters: Gemini is no longer the niche alternative many had written off after the rocky original Bard launch in 2023.

The Irony of a Full Reversal

The parallel with December 2022 is hard to ignore. Back then, it was Sundar Pichai who declared a code red inside Google, alarmed by how fast ChatGPT was capturing the public conversation and putting the search business that underpins Alphabet at risk. Google took months to respond with a competitive product and suffered very public stumbles — most memorably, Bard's botched demo in February 2023 — before restructuring its entire AI division under DeepMind and launching the Gemini family.

Three years later, the roles have reversed. It's OpenAI now trimming diversification bets — advertising, commerce, proactive features — to concentrate on defending the ground that defined its early lead. The company that taught the industry to move fast is now moving fast to avoid losing what it built.

What This Means for the Industry

A code red isn't an existential crisis, but it is a signal that competition among the major AI labs has entered a tougher phase. For users, the likely near-term outcome is a more frequently updated ChatGPT, with less noise from satellite features and more investment in what sustains its base: response quality and model speed. For the rest of the industry, the takeaway is that there's no longer a single undisputed leader, and the race for mass-market generative AI — which many considered settled back in 2023 — is wide open again.