Beyond Suspicious Steps: Ontological Trust in Long-Horizon Agents
Long-horizon agents increasingly operate across many steps, tools, and observa- tions. In this setting, the relevant oversight question is not only whether each action is locally valid, but whether the evolving trajectory still corresponds to the task the user authorized. Drift can accumulate quietly: an agent may call the right tool with plausible arguments at every step, while its prefix moves…
We haven't written up this one. arXiv cs.AI has the full story — the link below goes straight to it.
