OpenAI's Q2 Sales Growth Lags Rival Anthropic, Raising Concerns Over AI Valuation Bubble

The prevailing assumption across Wall Street has been that frontier artificial intelligence is a race to the bottom—a capital-incinerating contest where companies must subsidize users for years before seeing a dime of profit. The second-quarter financial disclosures from the industry's two biggest private players just shattered that consensus, but not in the direction most expected.[5]

OpenAI, long considered the undisputed leader of the generative AI boom, reported that its second-quarter revenue grew by just 18% quarter-over-quarter to $6.7 billion. For a company priced for hypergrowth, the sequential deceleration has alarmed shareholders who expected the startup to maintain its dominant market trajectory.[1][4]

Compounding the sluggish top-line growth, OpenAI's operating losses widened significantly. The company reported a $12.3 billion operating loss for the quarter, up from $9.3 billion in the first quarter. This pushes the ChatGPT maker further away from profitability just as the broader market begins to demand sustainable unit economics.[1]

In stark contrast, rival Anthropic more than doubled its revenue over the same period, reaching $11.6 billion. This represents a staggering 14-fold increase from the $787 million the company recorded in the second quarter of 2025, marking a rapid acceleration in its commercial footprint.[1][6]

More importantly, Anthropic reported a positive adjusted operating profit of approximately $559 million. This marks a historic milestone for the sector, challenging the entrenched thesis that leading AI models must remain money-losing loss leaders to capture market share.[5][6]

The divergence between the two AI giants stems from a fundamental shift in enterprise adoption. While OpenAI has faced slower growth for its flagship ChatGPT product, Anthropic's Claude Code has gained massive traction among software developers and corporate IT departments.[1][3]

Anthropic has aggressively targeted the enterprise API market, focusing on reliability, extended context windows, and advanced coding capabilities. By mid-2026, data from corporate payment processors indicated that a larger share of U.S. enterprises were actively paying for Anthropic's services than OpenAI's.[3]

This financial decoupling is forcing a rapid reassessment of the so-called "AI valuation bubble." The bubble thesis rests on the idea that AI companies are raising capital at astronomical valuations without a viable path to generating free cash flow.[2][5]

OpenAI's current trajectory exemplifies this risk. The company has reportedly locked in an estimated $1.4 trillion in long-term spending commitments, including massive deals with Microsoft and Oracle. These commitments were made under the assumption of exponential, uninterrupted revenue growth.[1][3]

With OpenAI's growth rate decelerating on a smaller revenue base than its chief rival, those capital commitments now look increasingly precarious. The widening gap between revenue and compute costs raises serious questions about the sustainability of its $882 billion private market valuation.[3][4]

Conversely, Anthropic's newfound profitability is fueling unprecedented valuation targets. The company is reportedly preparing for an October 2026 initial public offering that could value the firm at up to $2 trillion.[2][6]

If realized, an Anthropic IPO at that scale would eclipse SpaceX's recent listing to become the largest public debut in history. Investors are modeling the valuation on forward projections that Anthropic's annualized revenue run rate could reach $100 billion to $120 billion by the end of 2026.[2][6]

Because both OpenAI and Anthropic remain private, the immediate market impact is rippling through their publicly traded proxies and infrastructure partners. Microsoft, which holds significant exposure to OpenAI through investments and Azure cloud workloads, faces heightened scrutiny over whether its AI capital expenditures will yield the expected returns.[3][4]

Meanwhile, Anthropic's success is validating the massive data-center investments made by its primary backers, Amazon and Alphabet. The ability of an AI lab to generate positive operating income makes the hyperscalers' AI-linked debt and infrastructure spending look far more defensible to skeptical shareholders.[5]

The critical unknown moving forward is whether Anthropic's profitability is a durable structural advantage or a temporary artifact of its accounting practices. The company's adjusted operating income excludes certain stock-based compensation, and the underlying economics of its custom silicon partnerships remain opaque.[1][5]

Furthermore, the sheer scale of capital required to train the next generation of frontier models means that even a profitable quarter does not eliminate the need for massive ongoing fundraising. Both companies are locked in an arms race that demands tens of billions of dollars in specialized semiconductors and energy infrastructure.[2][4]

Ultimately, the second-quarter results signal that the AI market is maturing past its initial hype cycle. Enterprise customers are demonstrating a willingness to pay premium prices for high-quality, reliable models, but they are increasingly selective about which platforms they adopt.[1][6]

For investors, the grace period for cash-burning growth appears to be ending. The AI sector is no longer a monolith; it is dividing into platforms that can translate compute into sustainable cash flow, and those that are trapped subsidizing their own expansion.[3][5]