The authors built four environments in which simulated users each hand a task to an agent and the agents draw on one resource that cannot satisfy everyone: an API-token budget, a clinic calendar, a group order or booking, and a merge…
In the News: November 12, 2026 (Midday)
An Anthropic Fellows study finds one agent per user does worse than one shared agent in four test environments, a merge queue among them.
Midday edition
Machine-readable
Download Markdown
Story
One agent per user did worse than one shared agent in every environment an Anthropic Fellows study tested
Read story →
Story
Students reading code with an AI assistant carried more cognitive load and did not report it
Twenty students from two Java courses each read six Java snippets of about 30, 45 and 70 lines, half with a Gemini chat window open on a second monitor and half without, then answered four comprehension questions per snippet without the…