headline: OpenAI pivots to business customers as Anthropic rivalry intensifies slug: 2026-04-16-openai-pivots-to-business-customers-a…

  • OpenAI said Thursday it is redirecting resources from consumer products toward corporate clients and a forthcoming enterprise-grade AI mo…
  • Both OpenAI, valued at $852 billion, and Anthropic, valued at $380 billion, lose more money than they generate, placing the two San Franc…
  • “You’ll see a new model coming from us in short order. We feel very excited about it,” Friar said in an interview with The Associated Press.
  • OpenAI boasts more than 900 million weekly users of its ChatGPT product. Friar said about 95 percent of those users pay nothing, making t…

OpenAI said Thursday it is redirecting resources from consumer products toward corporate clients and a forthcoming enterprise-grade AI model, citing sharpening competition with rival Anthropic for business accounts. Chief Financial Officer Sarah Friar said business customers now account for about 40 percent of OpenAI’s revenue, up from roughly 20 percent when she joined the company in 2024, and she expects them to reach half of all sales by year’s end. The shift arrived the same day both companies released new AI models, escalating a race that each firm says it is winning.

Both OpenAI, valued at $852 billion, and Anthropic, valued at $380 billion, lose more money than they generate, placing the two San Francisco-based AI research labs under mounting pressure to convert free or low-cost users into revenue-bearing enterprise clients before their cash positions force a reckoning.

“You’ll see a new model coming from us in short order. We feel very excited about it,” Friar said in an interview with The Associated Press.

Pruning the consumer side

OpenAI boasts more than 900 million weekly users of its ChatGPT product. Friar said about 95 percent of those users pay nothing, making the free-tier operation a financial burden even as it builds brand familiarity across households.

The company discontinued its AI video generator app Sora to concentrate computing capacity on the new enterprise model. Friar described the decision as difficult but strategically necessary.

“I think it was a little heartbreaking, but we’re like, OK, it’s not the main event right now,” she said. “We need to make sure that our new model that’s coming has enough compute.”

Friar said the broader discipline applies to technology companies at scale. “Great companies are very good at, in a reasonable period of time, kind of doing that winnowing down and refocusing and it’s super painful,” she said.

New models on both sides

OpenAI is developing a model codenamed Spud, which the company describes as its “smartest model yet,” offering “stronger reasoning, better understanding of intent and dependencies, better follow-through and more reliable output in production.” Spud is positioned as OpenAI’s answer to Anthropic’s recently released Claude Mythos.

Anthropic claims Mythos is so capable in cybersecurity tasks that it has restricted access to select customers, citing the model’s apparent ability to surpass human experts in finding or exploiting computer vulnerabilities. Also on Thursday, Anthropic released Opus 4.7, described as its most powerful generally available model.

Hours after Anthropic’s announcement, OpenAI introduced GPT-Rosalind, named after scientist Rosalind Franklin, a specialized model designed to advance drug discovery and other life sciences research.

Reorienting around enterprise

Three months ago, OpenAI hired Denise Dresser, then chief executive of Slack, as its first chief revenue officer. Dresser said in a recent AP interview that she has focused on positioning OpenAI as the platform of choice for workplaces deploying AI agents to automate computer-based tasks.

“It’s really clear to me that companies are past the experimentation phase and they’re into using AI to do real work,” Dresser said. “Leaders at companies are recognizing that AI is probably the most consequential shift of their lifetime.”

Dresser sent a memo to OpenAI employees on Sunday, first reported by The Verge, that acknowledged Anthropic’s focus on software developers “gave them an early wedge” but asserted that OpenAI holds structural advantages as AI use expands beyond that base.

“Their story is built on fear, restriction, and the idea that a small group of elites should control AI,” Dresser’s memo said. “Our positive message will win over time: build powerful systems, put in the right safeguards, expand access, and help people do more.”

The revenue contest

Anthropic said its annualized revenues reached $30 billion. Friar and Dresser declined to disclose OpenAI’s current revenue but both suggested Anthropic’s figure is inflated because it does not account for revenue the company shares with cloud computing providers Amazon and Google, both major Anthropic investors.

Luke Emberson, a researcher at nonprofit institute Epoch AI, said the available data suggest the two companies are closer in revenue than public claims indicate — and that the gap may be closing.

“They’re likely quite close,” Emberson said. “Certainly the trends show Anthropic is growing much faster than OpenAI. If that continues, they’re likely to cross soon.”

Anthropic was founded in 2021 by a group of former OpenAI executives who said they wanted to prioritize AI safety. It has built a substantial following among software developers and positioned itself as the more safety-conscious vendor in the enterprise market.

Financial durability questions

For critics of the AI industry’s financial model, the trajectory of both companies raises concerns about the businesses that have come to depend on their services. Anthropic has imposed rate limits on heavy users, and both companies have set up service tiers that reward premium payers.

Ed Zitron, an author and AI industry critic, said the pattern follows a dynamic he finds familiar.

“It’s what I call the subprime AI crisis,” Zitron said. “People built their lives and they built their businesses on top of these companies that, as they try and save money, will start turning the screws.”

Zitron also challenged the premise that public stock listings would resolve the companies’ structural cost challenges.

“People will say, well, ‘Once they go public, they’re safe.’ That’s not true,” he said. “Public companies can and will die, especially ones that are dependent on $100 billion to $200 billion every year or so, just to keep breathing.”