Fetching from the wire…
Agents2026-09-16 · source-backed
CAC maps a typed action and its modeled risk to obligations naming predicates, evidence classes, scope, freshness and witness-set constraints. A deterministic evaluator returns satisfied, violated or unresolved, where unresolved emits a targeted evidence-acquisition request, and admission yields a certificate binding the action to its witness manifest. In 390 CAC trials, 120 effects completed with no harmful effects, while a live-policy baseline matched the completion count but admitted a constructed correlated-witness failure. The authors are explicit these are implementation behaviors, not production failure rates.
Each link below shares sources, entities, or timing with this story.
arXiv 2607.26598 targets the failure where an agent recovers from an error within an episode but hits the identical failure in later tasks, because post-episode feedback never revises the persistent harness. Guided by a domain-level Evolution-SOP, it writes episodic memory rec...
arXiv:2606.07889 names a failure mode where a coding agent holds information that should change its behavior, states that information out loud, and then acts against it anyway. The authors propose detecting this in execution trajectories as a pre-failure signal. For anyone run...
Li, Huo, and Johnson show that one-way message flow between agents produces neither mimicry nor solo behavior but an entirely novel dynamical state, at identical temperature settings. It's conceptual rather than quantitative, but the implication for orchestrator-worker fan-out...
Thirteen authors ran a generational genetic algorithm over specialized agents that separately handle mechanistic argument, assumption reconsideration, and evidence and testability assessment (arXiv 2609.15938). Evaluated against DepMap and Open Targets across 34 cancer types,...
Researchers loaded five systems with a revoked policy and its replacement, then measured retrieval and downstream action across nine policy scenarios, nine models and six defense conditions. Wherever the revocation label was visible to the retrieval layer, the revoked fact cam...
arXiv 2608.26733 presents an execution-only attack that reconstructs a hosted agent skill without ever asking the victim to reveal it, submitting crafted but ordinary tasks whose results discriminate between candidate hidden behaviors. At the weakest access level, final respon...
MindPattern daily
One email a day at 7 AM. Sources and a take on every story. Unsubscribe anytime.