Anthropic has told investors that its main focus is being a foundation AI model company, while leaving room to compete with application-layer startups. According to Tech.eu, the company said this may happen "from time to time" in "select categories". The statement comes from its S-1 filing in the US, published ahead of what the outlet calls one of the biggest IPOs ever.
What the filing shows
The figures reported from the filing:
- Revenue in the first half of 2026 was $16.3bn, against $1.2bn in the same period of 2025.
- Losses in the first half of 2026 were $80.8bn, against $13bn a year earlier.
- The latest private round, in May 2026, valued Anthropic at $965bn.
- No single customer accounted for more than 10 per cent of revenue in the nine months ended September 30, 2026.
Customer risk and competition
The filing also warns that a "significant portion" of Anthropic's customers are startups and technology-oriented enterprises. It says these may be more sensitive to adverse macroeconomic conditions and may cut or delay spending in a downturn.
On competition, Anthropic named OpenAI, Google and Meta. It said its focus on enterprise customers, its ability to "remain and define the frontier", and its talent set it apart.
What is not in the source
The source does not say which categories Anthropic might enter, when, or with what products. The statement is a general disclosure in an IPO document, not a product announcement. Tech.eu frames it as a long-standing worry for European application-layer startups, which fear that model providers could launch competing products and damage their businesses.
Why it matters for buyers
The filing confirms that a model provider with enterprise customers does not rule out selling applications directly. For a company that has bought AI tooling from a smaller vendor, that vendor's roadmap now depends partly on a supplier that may become a competitor. The same filing flags that startup customers could cut spending in a downturn, which is a reminder to check the financial resilience of the smaller AI vendors you rely on.