Anthropic4 Sept 2026 · 1 min · OriginalFermat’s Last Theorem now has a complete computer-checked proofClaude spent 11 days translating a monumental mathematical result into something Lean could verify, drawing on decades of human work.
Anthropic1 Sept 2026 · 1 min · OriginalSecurity monitoring with the records under customer controlAnthropic’s proposed safeguards separate who stores sensitive activity from who detects and reviews misuse.
Anthropic1 Sept 2026 · 2 minFable 5.1’s injection defenses include a fallback modelThe system card’s strong result measures a configured service, including interventions that change which model finishes the task.
Anthropic27 Aug 2026 · 1 min · OriginalA shared interface for laboratory equipmentAnthropic and HHMI Janelia are testing a standard that lets agents inspect device state and operate scientific instruments.
Anthropic10 Aug 2026 · 1 min · OriginalA substantial step on the Riemann hypothesis, with a precise limitAn unreleased Claude model raised a longstanding mathematical lower bound from 41.6% to 67.2%. The distinction between that result and solving the famous conjecture is essential.
Anthropic27 Jun 2025 · 1 min · OriginalThe AI shopkeeper could find tungsten cubes but could not reliably protect its marginProject Vend exposed a conflict between being helpful and running a business.
Anthropic26 Jun 2025 · 1 min · OriginalInstalling a local AI tool should not require becoming its developerDesktop Extensions addressed the last mile between an MCP demo and ordinary use.
Anthropic13 Jun 2025 · 1 min · OriginalResearch agents buy breadth with a much larger token budgetAnthropic’s multi-agent account explains where parallel work helps—and what it costs.
Anthropic8 Apr 2025 · 1 min · OriginalA student’s chat history cannot by itself tell you whether learning happenedAnthropic’s first education report separates observed requests from educational outcomes.
Anthropic20 Mar 2025 · 1 min · OriginalA deliberately uneventful tool gave an agent room to reconsiderThe “think” tool is also a lesson in how quickly AI engineering advice ages.
Anthropic10 Feb 2025 · 1 min · OriginalThe first Claude usage index measured tasks, not disappearing occupationsA foundational dataset is most useful when its sampling limits remain attached.
Anthropic6 Jan 2025 · 1 min · OriginalA coding benchmark score belongs to the model and its tools togetherAn early SWE-bench account shows how interface details affect measured capability.
Anthropic19 Dec 2024 · 1 min · OriginalAn agent architecture begins with deciding who controls the next stepAnthropic’s 2024 guide distinguishes fixed workflows from systems that choose their own path.
