Startup Anthropic is adding a series of upgrades to its Claude AI model — including the ability to interact with computer interfaces much like a human does. That ability, which the company calls simply "computer use,

Anthropic is making the agentic abilities available through its application programming interface (API) for its most powerful multimodal large language model, Claude 3.5 Sonnet, from today. The company also announced a new and improved version of a smaller model, Claude 3.5 Haiku, today.

Anthropic says it is teaching its Claude AI model to perform general computing tasks based on prompts. In demonstration videos, the model is shown controlling the cursor of a computer to conduct research for an outing on the town, searching the web for places to visit near the user’s home and even adding an itinerary to their desktop calendar.