Anthropic wants you to build on Claude. There’s a catch.

This piece looks at the unease among developers who build products on top of Anthropic's Claude models. Founders fear that the company supplying their core technology could move into their markets, using visibility into which integrations customers rely on. The article points to Spellbook, a legal AI startup, which says Anthropic launched a contract-review product that overlaps with its own business after assurances it had no such plans. Spellbook still uses Claude and is even featured as an Anthropic customer, showing the uneasy position of developers who rely on a platform that may also compete with them. The reporting notes that the evidence is largely anecdotal: no proof has surfaced that Anthropic uses integration data or proprietary partner information to guide its product decisions. Anthropic's privacy and commercial terms address some data use but do not clearly explain how aggregated integration data is handled. The outlet said Anthropic had not responded to its request for comment.