Anthropic Acquires Vercept to Expand Claude Computer-Use Stack
Original: Anthropic acquires Vercept to advance Claude computer-use capabilities View original →
What Anthropic announced
On February 25, 2026, Anthropic announced that it acquired Vercept to advance Claude’s computer use capabilities. In Anthropic’s framing, computer use is the ability for Claude to act inside live software interfaces and complete multi-step tasks that are difficult to solve with code-only workflows.
Anthropic said Vercept was built around hard perception and interaction problems, and that this expertise maps directly to the next phase of Claude’s agentic product roadmap. The Vercept team, including co-founders Kiana Ehsani, Luca Weihs, and Ross Girshick, will join Anthropic. Vercept’s external product is expected to wind down in the coming weeks.
Why this is tied to recent model progress
The company connected the acquisition to its recent Sonnet 4.6 launch. Anthropic said its Sonnet family improved on OSWorld, a widely used computer-use benchmark, from under 15% in late 2024 to 72.5% today. Anthropic also said this places the model near human-level performance on tasks such as spreadsheet navigation and multi-tab web form completion.
Strategic signal
The announcement suggests Anthropic is investing in both model intelligence and interface-level execution so Claude can handle real operational workflows, not only text generation. Anthropic described this as the latest team acquisition following Bun, and emphasized the same safety-and-rigor standard for integration.
Primary sources: Anthropic announcement and the original X post.
Related Articles
Anthropic has acquired Seattle-based AI startup Vercept to enhance Claude's computer use capabilities, folding the startup's desktop control technology and team directly into Claude development.
Why it matters: persistent memory is one of the missing pieces between demo agents and useful long-running agents. Anthropic pushed the feature into public beta on April 23 and framed it as a memory layer that learns from every session.
Anthropic’s new agent-market experiment matters because it turns model quality into money. In a 69-person office marketplace, Claude agents closed 186 deals worth just over $4,000, and Opus-backed users got better prices without noticing.
Comments (0)
No comments yet. Be the first to comment!