A new paper formalizes "authorization drift" — LLM agents exceeding their authorization boundaries while pursuing user goals without an adversary — across data, tool, scope, and time, and introduces the Authorization Drift Benchmark (ADB) with 648 samples across seven domains. Each ADB sample pairs a sufficient task with a boundary condition, testing whether agents stay within permitted limits during multi-step tool use.
LLM agents can exceed their authorization boundaries while pursuing user goals without an adversary. We formalize this phenomenon as authorization drift across data, tool, scope, and time, and introduce the Authorization Drift Benchmark (ADB): 648 samples across seven domains, each pairing a suffici
EU-based hosting, git-push deploys, automatic HTTPS, no cold starts. Free tier with a custom domain — perfect for shipping the agent you just read about.