What triggered the alarm
Two weeks before the memo, Google released Gemini 3 on November 18, 2025, posting record scores on benchmarks including MathArena Apex and beating OpenAI's models on several coding and reasoning tests. Google also disclosed that Gemini's monthly active users had grown from 450 million in July to 650 million by October, a figure repeated across the coverage of Altman's memo as the clearest sign of Gemini's momentum. The following week, Anthropic shipped Claude Opus 4.5, which it said beat Gemini 3 on coding and computer-use benchmarks, adding a second competitor to the pressure rather than just one. [1][3]
Atlas interpretation: The reporting around the memo did not cite a single ChatGPT metric that was falling in absolute terms. OpenAI was still citing 800 million weekly active users to Gemini's 650 million monthly figure, numbers measured on different windows that make a direct comparison unreliable rather than a clean story of ChatGPT losing ground. The more concrete signal came later: Fortune's follow-up reporting put ChatGPT's website visits at 5.844 billion in November, down from about 6 billion in October, a dip small enough to read as early softening rather than an overtaking, but enough to explain why Altman moved before the numbers got worse. [3][6]
What Altman told employees
The Information first reported the memo's contents on December 1, 2025, the day Altman sent it, and Fortune's account quotes the same core lines. Altman wrote that OpenAI was "at a critical time for ChatGPT" and told staff to prioritize the product's speed, reliability, and personalization. He warned employees to expect "temporary economic headwinds" from Google's resurgence and predicted, in his own phrase, "rough vibes" in the period ahead. [1][2]
Concretely, the memo pushed back OpenAI's advertising rollout inside ChatGPT along with three other efforts: Pulse, a proactive personalized-assistant feature; a shopping agent; and a health agent. Staff were told they could be temporarily reassigned onto ChatGPT work, and OpenAI instituted daily coordination calls to track the effort, details reported by The Information and corroborated in subsequent coverage from Fortune, SiliconAngle, and MacRumors. [2][3][4]
The same move, three years earlier
The headline comparison Fortune drew is accurate, not just a flourish. The New York Times reported on December 21, 2022, that Google CEO Sundar Pichai had declared his own code red weeks after ChatGPT's public launch, reassigning staff across research, Trust and Safety, and other groups toward a target of shipping competing AI prototypes and products by Google's I/O conference the following May. Pichai later said in a September 2023 interview that Google had held its own chatbot back because it "still had a lot of issues" with quality; the resulting product, Bard, was rebranded Gemini and launched in December 2023 to criticism over inaccurate outputs. [1][5]
Atlas interpretation: The symmetry is close to exact: same phrase, same three-year gap, same underlying fact that a leadership position in this market has changed hands twice in as many product cycles. Fortune's framing treats OpenAI's 2025 code red as evidence that the company ChatGPT displaced in 2022 has, at least by Google's own benchmark and usage numbers, closed the distance it opened three years earlier. [1]
What OpenAI did next, and how the moment read
OpenAI shipped GPT-5.2 in mid-December, followed on December 16 by a new image-generation model aimed at Google's Nano Banana Pro. Fortune's follow-up reporting, published December 17, quoted Fidji Simo, OpenAI's CEO of applications, describing the code red as a way "to really signal to the company that we want to marshal resources in one particular area, and that's a way to really define priorities," and COO Brad Lightcap saying enterprise adoption was the unfinished opportunity he focused on daily. The same reporting put the code red's expected length at eight weeks, longer than prior internal pushes because of the holiday period, with an internal target of winding it down in late January behind a second major product release. [6]
Atlas interpretation: The optics mattered as much as the substance. OpenAI had been the presumed leader since GPT-5 shipped four months earlier, and a public, reported internal alarm was a visible reversal of that posture, the kind of defensive move OpenAI's launches had forced on competitors rather than the other way around. Fortune's own enterprise-market figures complicate a simple leader/challenger story further: at the time of its December follow-up it put OpenAI at 27% of enterprise AI share against Google's 21%, with Anthropic ahead of both at 40%, meaning the ground Altman's memo was defending was ChatGPT's consumer position specifically, not a lead OpenAI already held everywhere it competes. [6]
Sources
- Sam Altman declares 'Code Red' as Google's Gemini surges—three years after ChatGPT caused Google CEO Sundar Pichai to do the same
Fortune · Dec 2, 2025
- OpenAI CEO Declares 'Code Red' to Combat Threats to ChatGPT, Delays Ads Effort
The Information · Dec 1, 2025
- OpenAI CEO Sam Altman reportedly declares 'code red' as rivals launch new LLMs
SiliconANGLE · Dec 2, 2025
- Sam Altman Declares 'Code Red' for ChatGPT, Delays OpenAI Advertising Plans
MacRumors · Dec 2, 2025
- Google at 'code red' over ChatGPT as teams reassigned to work on competing AI products
9to5Google · Dec 21, 2022
- Inside OpenAI's fragile AI lead, and the 8-week 'code red' to defend it
Fortune · Dec 17, 2025