Anthropic’s growth is heavily concentrated in volatile enterprise spending, such as R&D budgets. Large customers can delay, renegotiate, or abandon contracts altogether. Because of this, the company’s financial performance is highly sensitive to changes from their relatively small customer base.
More fundamentally, Anthropic’s cost structure complicates its bullish narrative. The company is a compute-intensive business, dependent on the construction of an incredible amount of infrastructure to turn a profit. The company has entered into multibillion-dollar arrangements with firms such as Microsoft and Nvidia, committing tens of billions of dollars to cloud and chip capacity to support its future models.
This raises an uncomfortable question for Anthropic: as revenue gets pushed further down the road, where precisely do the profit margins come from?
In industries where input costs are both substantial and controlled by a handful of powerful suppliers—who are themselves potential investors—economic rents tend to be competed away. Anthropic’s reliance on speculative investment from a small group of funders places it in a structurally weaker position than narratives suggest. It is difficult for it to claim a durable moat when its largest cost centers are owned by firms with both the incentive and the capital to replicate its product.
This is particularly salient given the competitive AI landscape. Google’s DeepMind, Meta, and OpenAI are collectively investing tens of billions into similar capabilities. In such an environment, differentiation must be rooted in something more durable than branding or positioning.
Even the company’s headline metrics warrant scrutiny. As Reuters reported, Anthropic’s widely cited “run-rate” revenue figures are based on extrapolations of recent usage rather than realized annual sales, with cumulative historical revenue reportedly closer to $5 billion than the reported $30 billion through 2025. This is not unusual for high-growth firms, but it underscores how much current valuations depend on assumptions about the future rather than performance to date.
Anthropic CEO Dario Amodei recently said that “the pressure to survive economically while also keeping our values is just incredible.” Mission-driven companies often struggle in public markets precisely because their objectives are not easily reducible to quarterly earnings.
Governance is where Anthropic’s story looks especially fragile. Its bespoke structure, including the Long-Term Benefit Trust, is explicitly designed to blunt shareholder influence and shield management from near-term accountability. That may appeal to a mission-driven culture, but it asks public investors to bankroll a company they cannot realistically control.
In practice, this means the people taking the economic risk will have the least say over strategy, capital allocation, or leadership, even as the company pursues one of the most aggressive valuations in market history. Public markets occasionally tolerate that kind of arrangement when the cash flows are undeniable. When they are not, they tend to punish it quickly.
The risk for Anthropic is that its valuation assumes a level of economic performance that even successful technology companies rarely achieve. When that gap becomes apparent—as it often does post-IPO—the adjustment is usually swift and unforgiving.
