OpenAI’s $80 billion valuation questioned: Holding 1 billion users but shifting to B2B, forced into a corner by competitors?

```

OpenAI, whose valuation set a global unicorn record, is facing unprecedented internal skepticism from investors.

According to the Financial Times, some early investors in OpenAI have publicly questioned its $852 billion valuation. The core issue lies in the company’s urgent rush to pivot to the enterprise market—even as it owns ChatGPT, a consumer product with over one billion users—but finds itself completely suppressed by competitor Anthropic in this most critical arena. At the same time, a rare situation has arisen in the secondary market where investors are vying to buy Anthropic shares while snubbing OpenAI’s stock.

The anxiety caused by strategic vacillation has now spread to the realm of capital. Reports indicate that insiders say that justifying investment at OpenAI's current round would require assuming an IPO valuation above $1.2 trillion—a premise that is hard to sustain now that Anthropic offers a lower-priced alternative at a $380 billion valuation. OpenAI’s CFO Sarah Friar rebutted this, emphasizing that the just-completed $122 billion fundraising was oversubscribed and set a historic record, and that claims "investors don't support our strategy" do not match the facts.

Valuation in doubt, Secondary shares Cooled

As OpenAI completed the largest private fundraising in history in the primary market, the secondary market sent signals in the opposite direction.

According to Bloomberg, multiple institutional shareholders attempted to sell approximately $600 million of OpenAI stock on secondary market platforms, but after reaching out to hundreds of institutional buyers, none took the deal. In contrast, the same group of buyers is reportedly sitting on about $2 billion in cash, waiting specifically to buy Anthropic shares.

On special purpose entity trading platforms such as Augment and Hiive, the trend toward buyers favoring Anthropic is also apparent: OpenAI's secondary market valuation for legacy shares is about $765 billion, a discount of about 10% compared to the previous round; while Anthropic's secondary market valuation has reached $600 billion, representing a premium of over 50% on the last round.

An investor holding shares in both companies said that supporting this round’s investment in OpenAI would require assuming an IPO valuation of over $1.2 trillion. "Anthropic is cheaper to buy, and OpenAI risks being trapped in no-man's-land."

Iconiq Capital partner Roy Luo stated directly: "Fundamentally, the market has a first-place/second-place dynamic, and the number one brings disproportionately high returns. We've made our choice and invested more than $1 billion in Anthropic." He added that the current shift in sentiment isn’t definitive: "Last year, everyone was congratulating OpenAI's early investors."

Falling Far Behind on B2B, Strategy Switching Frequently

The rapid growth of Anthropic's annualized recurring revenue was the direct trigger for this round of doubts. Reports indicate Anthropic’s annualized revenue surged from $9 billion at the end of 2025 to $30 billion in less than four months, surpassing OpenAI’s previously disclosed $25 billion.

OpenAI’s Chief Revenue Officer Denise Dresser claimed in an internal memo that Anthropic had "inflated revenue by about $8 billion" from revenue-sharing with Amazon and Google. Anthropic responded that it is the principal in such transactions via partner channels, and that recognizing revenue based on gross amount is in line with industry accounting standards. Both companies claim to follow standard practices, but the different methods make direct comparison difficult.

Nevertheless, the dispute over revenue metrics cannot cover up OpenAI’s structural disadvantages in the enterprise market. Dresser admitted in the memo that Anthropic's "focus on code gave them an early foothold in competing for enterprise clients," but emphasized, "We still have a chance in this market."

Meanwhile, OpenAI’s strategic initiatives have noticeably accelerated, but internal critics see this as reflecting an unclear direction. CEO Sam Altman issued a "red alert" to employees at the end of last year, calling for focus on core business. Applications CEO Fidji Simo also urged employees to abandon "side tasks."

Yet just two weeks later, the company spent a "low hundreds of millions" to acquire the tech talk show TBPN, leaving some investors puzzled. "Frankly, I don’t understand it—it makes no sense to me, it’s a distraction and it annoys me," an OpenAI investor said.

The distractions for investors continue: Video generation service Sora has been shut down, resulting in the collapse of a previously negotiated $1 billion deal with Disney; Microsoft has announced it will take legal action if OpenAI’s new $50 billion partnership with Amazon violates its exclusive cloud agreement; Stargate data center plans have also been sharply scaled back, the $3 billion UK data center project and the Texas expansion have both been shelved, and a multibillion-dollar partnership with Nvidia has been drastically cut.

Sapphire Ventures President Jai Das compared OpenAI to the "Netscape of the AI era"—once dominant but eventually surpassed by later entrants.

Computing Power Reserves as Key Moat

Amid many strategic uncertainties, OpenAI’s infrastructure accumulation remains a significant advantage.

According to company disclosures to investors, OpenAI has now secured 8 GW of compute power resources, claiming that Anthropic won’t reach this scale until the end of 2027; OpenAI’s goal is to expand compute reserves to 30 GW by the end of 2030. Anthropic has not yet released its expansion plans, with CFO Krishna Rao saying the company will remain "disciplined" in its infrastructure build-out.

An individual involved in OpenAI infrastructure said: "Even if our model is slightly inferior, we can guarantee stable service." Anthropic, on the other hand, has suffered multiple outages and power bottlenecks.

OpenAI is now directing compute resources toward enterprise products, focusing on driving enterprise sales for its code generation tool Codex. Several strategic insiders expect Codex could surpass ChatGPT in prioritization. "This is a much higher margin business; technically, relocating compute from consumer to enterprise isn’t complicated," said one person.

The company plans to expand its employee count to 8,000 by the end of the year, by which time the proportion of enterprise revenue is expected to rise from the current 40% to 50%. Sarah Friar stated the $122 billion in fundraising "gives us immense flexibility at this moment," providing ample room for strategic choices.

Consumer Moat Depth in Question

ChatGPT remains the world’s largest AI app by user numbers, but the efficiency of monetizing this advantage is being questioned.

An early OpenAI investor told the Financial Times: "You have ChatGPT, a business with a billion users growing 50% to 100% a year, and you're talking about enterprise and code? This is an extremely unfocused company."

The vast free user base on the consumer side consumes immense compute power, but generates very little revenue. At the same time, OpenAI’s moves such as shutting down Sora and cutting multiple long-term research projects have also led to concerns about its continuity in innovation.

Another major investor holds a different view: "The company used to bet on too many directions. Now it needs to refocus on several core bets. You can’t compete on 30 fronts at once."

OpenAI launched ChatGPT ahead of the world in November 2022, igniting the global AI investment boom and successfully transforming itself from a non-profit organization into a commercial company. But now, with secondary market premiums flowing to its competitors and investors questioning its strategic coherence, the company that once defined the era of AI single-handedly now faces a test of focus and trade-off.

Risk warning and disclaimerThe market carries risk and investments should be made with caution. This article does not constitute personal investment advice and does not consider any user's special investment objectives, financial situation, or needs. Users should consider whether any opinions, viewpoints, or conclusions herein are suitable for their particular situation. Invest at your own risk. ```