In 2025 alone, OpenAI released a controversial text-to-video generator, dubbed Sora, and an abysmally slow web browser called Atlas. It also announced top-secret hardware alongside former Apple exec Jony Ive, and signed a $200 million contract with the US Department of Defense.

Meanwhile, the company continues to burn through billions of dollars a month, astronomical losses that have executives there feeling agitated. The company recently announced that it’s planning to spend a whopping $600 billion on AI infrastructure by 2030, an ungodly sum only beaten out by its original promise: $1.4 trillion, more than twice the revised figure.

Now, the company is coming to terms that it may have spread itself too thin, and is now looking to refocus its resources on its coding and enterprise users.

As the Wall Street Journal reports, OpenAI’s CEO of applications, Fidji Simo, told employees that the company is “actively looking at which areas to deprioritize.”

“We cannot miss this moment because we are distracted by side quests,” she told staff, as quoted by the WSJ. “We really have to nail productivity in general and particularly productivity on the business front.”

The move suggests the Sam Altman-led company is really starting to feel the pressure as the walls continue to close in, with spooked investors questioning when — or if — they’ll ever benefit from digging deep into their pockets to fund the venture.

The news also shows that the ChatGPT maker is watching as rival Anthropic continues to make headway, quickly establishing itself as the enterprise-facing AI company to beat, following the runaway success of its Claude Code software.

The momentum of Claude Code — which, alongside its agentic AI assistant Claude Cowork, triggered a trillion-dollar selloff last month amid fear that they could make traditional software-as-a-service companies obsolete — is palpable.

OpenAI, on the other hand, has thrown lots of spaghetti at the wall, dabbling in image and video generators, among other side projects. Current and former employees told the WSJ that OpenAI lost much of its focus last year.

And the stakes continue to grow as AI companies, including OpenAI, Anthropic, and Elon Musk’s xAI (now technically under SpaceX’s auspices), are rumored to go public later this year.

AI companies are also continuing to fight over dwindling access to computing power, with data center capacity becoming increasingly harder to come by. It’s an especially complicated dance when multiple projects are involved. OpenAI staffers told the WSJ that the company’s organizational structure has turned into a mess.

OpenAI’s latest woes are strongly reminiscent of Altman declaring “code red” last year, with Google’s Gemini emerging as a very real threat. At the time, Altman urged staffers to improve the quality of the company’s blockbuster chatbot, even at the cost of delaying other projects.

Given the latest news, the warning lights have yet to turn off after blinking for three months straight at OpenAI’s headquarters.

“We are very much acting as if it’s a code red,” Simo told staffers, using the same language Altman did in December. “I don’t think necessarily declaring codes for everything makes a ton of sense.”

More on OpenAI: Details Emerge About OpenAI’s “Adult Mode”