ALE

Top AI agents scored 0% on expert-level professional tasks in the ALE benchmark, with models like Fable 5 and GPT-5.5 failing to solve any problems requiring real expertise. Mid-level performance reac…

Researchers introduced Agents' Last Exam (ALE), a new benchmark designed to evaluate AI agents on long-horizon, economically valuable, real-world tasks with verifiable outcomes. Developed with over 25…

Agent Lineage Evolution (ALE) framework, arguing that AI agent sessions must be regularly terminated and restarted to prevent context pollution and reasoning degradation, a process requiring human sup…