News

Inside OpenAI's Code Red Moment

As Google's Gemini surges, Sam Altman pulls the plug on distractions and doubles down on ChatGPT. But is it too late?

• By John K. Waters
• 12/07/2025

In December 2022, Google declared a "code red." ChatGPT had just launched and was sucking up the internet's attention like a black hole. Google's search empire, suddenly vulnerable, pivoted hard to AI. Teams were reshuffled. Prototypes rushed. Sundar Pichai himself sounded the alarm.

Three years later, the tables have turned. Sam Altman is now the one waiving a red flag.

In a leaked internal memo this past week, the OpenAI CEO issued a "code red" to employees. The message: Stop what you're doing. Refocus on ChatGPT. Kill or delay anything that's not making the world's most popular chatbot better, faster, or more useful—now.

At first glance, it might seem like just another CEO rallying the troops. But for OpenAI—a company long admired (and criticized) for its sprawling ambitions—the directive is something more: a strategic about-face in the middle of a high-stakes race.

Google's Gemini 3 is gaining traction, toppling OpenAI models on benchmark tests and steadily closing the user gap. Altman's urgent memo signals a company trying to reassert control not just over the market, but over its own identity.

"We are at a critical time for ChatGPT," Altman wrote in the memo, pushing back work on advertising, health agents, shopping assistants, and a personal assistant named Pulse. Teams are being reassigned. Daily calls have been mandated. The clock is ticking.

The Fragmentation Problem
OpenAI was never built for focus. Unlike Google, whose early mantra—"organize the world's information"—locked in a singular mission, OpenAI spread its bets. It shipped code generation tools, business APIs, custom GPTs, image generators, and half-baked product experiments. The company called ChatGPT a "happy accident," a side effect of showing off what its models could do.

But the past year has shown just how costly that scattershot strategy can be.

Gemini 3, Google's latest model, launched in November with better performance on key benchmarks and a slick integration into the world's most trafficked search engine. It now boasts 650 million monthly users, up from 450 million just months earlier. Marc Benioff, Salesforce CEO and former ChatGPT evangelist, publicly switched allegiances, calling Gemini's leap "insane." Meanwhile, OpenAI's GPT-5 landed with a thud in August, criticized for being too stiff, too cautious, and frustratingly bad at things like geography and math.

In short: ChatGPT lost its edge. Now OpenAI is trying to get it back.

Altman's Bet: Fix ChatGPT or Die Trying
Altman's code red is part of a broader attempt to tighten the company's focus around a product that still draws 800 million weekly users but is no longer the clear innovation leader.

According to people familiar with the plan, the company is rolling out a new reasoning model this week that it claims outperforms Gemini 3 internally. Whether that's enough to shift momentum remains to be seen.

But the real question isn't whether OpenAI can build a better model. It's whether it can become a better company.

For months, OpenAI tried to do everything. Build consumer tools. Chase enterprise deals. Fund a startup ecosystem. Invest in chips and data centers. Launch weird little products like a video-based social platform. Meanwhile, its core user experience stagnated. Updates oscillated between too sycophantic (GPT-4o) and too robotic (GPT-5). Even OpenAI's own teams seemed unsure how ChatGPT should sound.

"The only thing bigger than the company's attention deficit is its appetite for capital," Reuters columnist Robert Cyran noted last week. OpenAI has committed more than $1 trillion in future obligations to infrastructure partners like Microsoft and Oracle. It's not yet profitable. And unlike Google, it can't subsidize its AI ambitions with ads.

Altman's decision to pause advertising efforts—at a moment when monetization pressure is peaking—reflects how urgently he views the threat.

The Danger of Playing Defense
There's an irony to all this, of course. The AI boom was kicked off by ChatGPT, a tool that wasn't even supposed to be a product. Now, it's the one thing holding OpenAI's sprawling empire together.

But refocusing on ChatGPT isn't without risk. By shelving Pulse, health agents, and other experimental features, OpenAI may be delaying the next breakout app. Innovation thrives on accidents—and ChatGPT itself was one.

That tension—between focus and experimentation—is one Altman will have to navigate carefully. The company's stated mission is to ensure that artificial general intelligence (AGI) benefits all of humanity. That's a noble ideal. But it doesn't say much about how to prioritize product features when Google is breathing down your neck.

For now, OpenAI seems to be hunkering down. Whether that's a moment of clarity or a sign of panic depends on what happens next.

The Road Ahead
Altman's memo suggests he's betting on a return to executional excellence. Make ChatGPT faster. Smarter. More personal. Less annoying. Make it feel like a product that's still improving, not coasting on past success.

Nick Turley, OpenAI's head of ChatGPT, posted that the team is now focused on making the experience "more intuitive and personal." The company has also revived its internal urgency color scale—from yellow to orange to red—to signal the severity of the moment.

There's little doubt the pressure is real. Anthropic is gaining traction among business users. Meta is scaling fast. Google has found a groove. And OpenAI, still private, must continually raise funds while convincing the world it's still the one to beat.

But in tech, as history shows, crises can be clarifying. Google found focus after its own "code red." Now it's OpenAI's turn to see what urgency can build—and what it might cost.