Thirty billion dollars raised. A pre-money valuation of 350 billion dollars, or approximately 297.5 billion euros. In a few years, Anthropic has risen to a level comparable to that of large international listed groups, without having yet entered the stock market. The question is no longer that of technological performance, but financial: to what fundamentals does such a valuation really correspond?
An anchor point: $14 billion run rate
The company indicates a run rate of 14 billion dollars (around 11.9 billion euros). Four-fifths would come from corporate customers. More than 500 organizations reportedly spend more than $1 million a year on its tools, including Claude Code, positioned as a central assistant in development environments.
This revenue profile is structuring, because it suggests significant penetration into critical workflows, therefore high retention potential, but it says nothing about margins.
Compared to the announced run rate, the implicit valuation represents approximately 25 times the annualized turnover. A multiple which far exceeds the current standards of mature SaaS and is more part of a logic of anticipation than of a snapshot of present fundamentals.
🚨 SMARTJOBS
- ECOLE POLYTECHNIQUE – Director/Deputy Director of International Relations (F/M)
- LEVELLR — Head of Sales (EMEA)
- CLAROTY — Sales Development Representative
- CURE51 — Data Scientist (Internship)
- FRACTTAL — Account Manager (France)
- ONE-FIVE — Product Owner / Product Manager
- BRICKSAI — Founding Growth Manager
👉 Find all our offers on the DECODE MEDIA Jobboard
đź“© Are you recruiting and want to strengthen your employer brand? Discover our partner offers
The multiple: future growth versus current profitability
A multiple of this order assumes either sustainable hypergrowth or high future margins, or both. Assuming a trajectory towards $40 to $50 billion in revenue over the next few years, the valuation could normalize mechanically. Growth must still be maintained and the cost structure must not erode profitability.
This is where the equation becomes more complex, because foundational AI models are not a pure software model and rely on a heavy, energy-intensive computing infrastructure dependent on specific actors.
The blind spot: the cost of computing
Access to Nvidia GPUs and data center capacities constitutes a central position. The amounts committed to the ecosystem, whether direct investments or cloud credits via partners like Microsoft, reflect an industrial reality.
Anthropic’s actual gross margin is not public, and depending on the scenario, the equation changes significantly. So if it approaches SaaS standards (70% or more), the multiple finds a form of consistency, on the other hand if the cost structure brings the model closer to a cloud infrastructure (30 to 50% gross margin), it is not the same story.
This information is important because it determines the very nature of Anthropic, is it a high value-added software publisher or a capital-intensive infrastructure operator?
Pressure on talent
Added to this material dimension is competition for specialized researchers and engineers. Remuneration, share packages and the scarcity of profiles constitute a high structural cost. In a context of direct competition with OpenAI, Google or Meta, the highly qualified payroll becomes a key factor of differentiation, but also of potential dilution. The balance between investment in R&D, commercial growth and cost control will be carefully scrutinized during a possible IPO.
Towards a full-scale test IPO
Anthropic is preparing to enter the public markets. The exercise will require increased transparency, namely the detail of margins, the level of capex, cash flows, contractual exposure. Public markets promote growth, but quickly punish uncertainty about profitability. And at $350 billion, the private market already includes part of the future promise, which means that the IPO will have to transform this promise into a credible trajectory.
A new asset class?
Beyond financial ratios, valuation arguably reflects a deeper shift, namely that foundational AI models are gradually being seen as a strategic infrastructure layer, comparable to the cloud or semiconductors. In this reading, the market does not only value a stream of income, but a systemic position in the digital economy.
This interpretation also explains the convergence of industrial investors, sovereign funds and venture capital in the round table.
Between technological bet and financial discipline
Anthropic’s financial equation is therefore based on a precise sequence which applies to OpenAi or Mistral AI, namely maintaining sustained growth, stabilizing margins compatible with a high multiple, securing access to computing and transforming its integration into companies into lasting dependence.
At $350 billion, the valuation is not disconnected in principle. It assumes that the company is not only an efficient laboratory, but an essential platform, capable of reconciling capital intensity and financial discipline. It is on this point that the coherence, or not, of the equation will be decided.
An oversubscribed tour
The $30 billion (around €25.5 billion) funding round brings together a group of leading financial and industrial players. It includes part of the $15 billion commitment announced in 2025 by Microsoft and Nvidia. Alongside these two strategic partners are the funds Coatue, Founders Fund, Sequoia Capital and Greenoaks, as well as the sovereign investors GIC and Temasek. Other backers include Altimeter Capital and MGX. According to sources close to the matter, the portion not covered by the Microsoft / Nvidia commitment would have been several times oversubscribed, leading to an increase in the initial fundraising objective.
