What does OpenAI’s ‘Code Red’ warning mean for Microsoft?

Three years ago, OpenAI jumpstarted the generative AI revolution when it released ChatGPT, and the world hasn’t been the same since. Hundreds of billions of dollars have been invested in AI, OpenAI founder and CEO Sam Altman became the face of AI, and billion-dollar AI companies seem to be minted at will.

As an early investor in OpenAI, Microsoft has been changed as much as any company by the genAI revolution. The tech giant’s value has skyrocketed to between $3 trillion and $4 trillion depending on the day, and it’s become the premiere AI firm in the world.

But now all that may change. On December 1 Altman sent out a memo to OpenAI employees declaring a “Code Red” emergency and focusing all company efforts on improving ChatGPT. The reason? Google’s newly released Gemini 3 model beat the pants off GPT-5.1, the current version of the large language model (LLM) behind ChatGPT, in just about every benchmark test.

It’s clear the future will be dire for OpenAI if Gemini continues to best ChatGPT. Not so clear are the effects on Microsoft, which uses OpenAI’s GPT-4 and GPT-5 models to power its own  Copilot chatbot. To better understand what they might be, we’ll first take a look at how Google’s latest Gemini release stacks up against ChatGPT.

Battle of the LLMs

Analysts point to a variety of ways in which Gemini 3 bests GPT. Mayank Kejriwal, a principal scientist at the USC Information Sciences Institute, says Gemini 3 has seen the greatest leap in capability this year, while GPT and other LLMs saw only “incremental updates.”

Kejriwal told Business Insider that the latest version of Gemini is well ahead of GPT on the LMArena leaderboard, in which users rate the accuracy and comprehensiveness of chatbots in a number of arenas including text, text-to-image, and search. Gemini 3 is number 1 overall, while GPT-5.1 is far behind at number 6.

Google has released a wide variety of specific benchmarks that show how Gemini has surpassed GPT-5.1. Altman’s declaration of a Code Red in response indicates that this is not mere hype.

And Google immediately embedded Gemini 3 across its core products, including search. Sanchit Vir Gogia, chief analyst at Greyhound Research, said that integrating Gemini directly into Google Search is one of the most important shifts in the AI enterprise market in the last year.

“This is not an AI feature layered on top of search,” he told Computerworld. “It is a fundamental rewrite of the global information distribution engine that billions rely on every day. For enterprises, this marks a decisive moment where AI is no longer a separate capability but the default interpreter of user intent, workflow context, and knowledge retrieval.”

What does this mean for Microsoft?

It’s clear this is all bad news for OpenAI. But what does it mean for Microsoft?

For a start, the news about Google’s breakthrough isn’t as dangerous for Microsoft as it would have been if it had happened a year ago. Back then, Microsoft was almost completely dependent on OpenAI. But the company has taken big steps to augment and even replace GPT as its go-to genAI.

These intentions became clear when Microsoft hired prominent AI leader Mustafa Suleyman to head up its new AI division in March 2024. Suleyman is a co-founder and former head of Applied AI at the AI company DeepMind, which was bought by Google.

A year and a half later, Microsoft launched its first homegrown LLMs, which ultimately may augment or even replace GPT as Copilot’s brains. It also made a deal with OpenAI competitor Anthropic to power parts of Microsoft 365 Copilot, the enterprise version of the chatbot that integrates with Microsoft 365 apps. That deal fixes one of Copilot’s main problems — its Excel capabilities are exceedingly weak, while those of Anthropic’s Claude model are much better.

That’s just the beginning. Eventually, Microsoft may take a “best-of-breed” approach that relies on multiple LLMs, not just GPT, to power Copilot.

Suleyman recently announced a new direction for Microsoft’s AI plans that go well beyond today’s chatbots. He calls it “Humanist Superintelligence,” and it’s more than just an empty slogan. He says it will “solve real concrete problems and do it in such a way that it remains grounded and controllable. We are not building an ill-defined and ethereal superintelligence; we are building a practical technology explicitly designed only to serve humanity.”

In practical terms, that means not one-size-fits-all chatbots like ChatGPT and Gemini, but instead a series of AI-based technologies, each pointed at solving an important problem and aimed at bettering people’s lives. The first two to be rolled out will be for medicine and improving energy efficiency.

The Google enterprise threat

All that being said, Gemini’s emergence as the world’s premier chatbot does present potentially serious problems for Microsoft. Microsoft 365 Copilot, which requires an add-on subscription, hasn’t been adopted in enterprises the way the company had hoped. A year ago, Microsoft announced that almost 70% of Fortune 500 companies were using Microsoft 365 Copilot. But many of those companies were using it in pilot programs, not for enterprise-wide use.

It appears that may still be the case. Gartner has warned that 47% of IT leaders have no confidence or are not very confident in being able to manage security and access risk in Copilot. An MIT report, The GenAI Divide: State of AI in Business 2025, found that 95% of genAI pilots fail — and most enterprise genAI pilots are for Copilot.

If Gemini proves to be a big improvement over Copilot, enterprises may start turning to Google Workspace and abandoning Microsoft 365. That would have far more serious consequences for Microsoft than just lost revenue for one product. Once Google gains deep relationships with an increasing number of enterprises, it could supplant other Microsoft products in them as well, notably Azure. Google could then eventually overtake Microsoft as the world leader in AI.

The upshot

None of this is a done deal, of course. GenAI models typically leapfrog one another, so don’t be surprised if the next version of GPT — and therefore Copilot — will perform better than Gemini.

But it is the first time that Microsoft’s dominance in AI has been challenged. So it may well be time for Microsoft to announce its own Code Red.