Anthropic’s annualized sales have quintupled in five months, reaching nearly $50 billion, surpassing OpenAI to become the dominant seller of models to enterprise customers. But as the company expands its own application development, it’s now directly competing with its partners’ businesses — and those partners are getting worried.
Ahead of its April launch of the prototyping tool Claude Design, Anthropic had lined up companies including Figma and Canva as launch partners. But Claude Design introduced collaboration and advanced editing features that overlapped with the core products of both design firms. Figma pulled out days before launch, and Anthropic chief product officer Mike Krieger resigned from Figma’s board around the same time. Figma CEO Dylan Field later accused Anthropic of a lack of transparency during an industry event.
In the coding space, Anthropic’s Claude Code has already surpassed Cursor and Microsoft’s GitHub Copilot in revenue. Its recently released legal and financial tools are also going after the large customers of startups like Harvey, Rogo, and Legora. On the technical side, the newly launched Mythos model has an implicit performance degradation for tasks related to developing advanced AI hardware and software, raising developer concerns that core functionality might be compromised.
On pricing, Anthropic has shifted to usage-based billing and introduced a new tokenizer, causing costs for large customers to spike — some companies have already started limiting employee usage. Jeff Reihl, technology chair at LexisNexis, noted that usage-based billing makes costs unpredictable, and law firms in particular need predictability. To counter the competition, rivals are lobbying customers to adopt multi-model strategies or more cost-effective open-source alternatives.