arXiv:2607. 22868v1 Announce Type: new Abstract: Runtime guardrails act before irreversible tool calls, but their guarantees depend on what policy state is representable, what a judge observes, and whether intervention changes future behavior.
arXiv:2606. 28639v2 Announce Type: replace-cross Abstract: We establish the mathematical limits of AGI safety in two forms: verifying a fixed system, and verifying that a certified safety property persists once the system self-modifies.
arXiv:2607. 16895v1 Announce Type: new Abstract: Safe adaptive control is online adaptation under a safety guarantee on the learning trajectory itself.
arXiv:2606. 31023v1 Announce Type: cross Abstract: Hard-constrained sequential decision systems have no certified way to spend the test-time compute of modern AI: executing the multi-step drafts of a learned policy or a frozen LLM forfeits the feasibility guarantee a trusted solver provides, while invoking the solver at every step forfeits the speed the AI offers.
The paper investigates the safety of Code World Models, where a language model generates an executable world model that a planner uses, and the model is accepted if it reproduces sampled transitions. It defines the pipeline’s danger as the expected risk, showing that the probability of missing a critical event across N independent rollouts is (1‑r)^N, and that an additional acceptance sample adds to the exponent. Experiments on hybrid instruments reveal that mode‑blind models can be exploited, and the authors provide theoretical bounds on localization budgets and demonstrate that acceptance only guarantees sample consistency, covering about two percent of the planner’s queries.
The paper investigates the safety of Code World Models, where a language model generates executable world models that a planner uses. It shows that accepting a model based on sampled transitions only guarantees sample consistency, not full safety, because the probability of missing critical events decays as (1‑r)^N. Experiments on hybrid instruments reveal that omitted mode‑boundaries can severely limit planner performance, and that even sophisticated LLMs (GPT‑5.x) struggle to repair such omissions in higher‑dimensional settings.
The paper demonstrates that safety mechanisms for autonomous large language model agents fail to compose across iterative loops, as trajectory‑scoped monitors cannot detect attacks whose evidence is spread over multiple iterations. It introduces LoopHarness, a system that maintains a persistent, non‑decaying safety state across loops, bounding unauthorized actions with a constant that does not grow with the number of iterations. The authors provide a comprehensive evaluation protocol, including attacks that require cross‑iteration evidence, module ablations, and adaptive white‑box red‑team testing.
arXiv:2608. 12700v1 Announce Type: new Abstract: Systems that generate GPU kernels with language models report high correctness rates.
arXiv:2607. 13070v1 Announce Type: cross Abstract: Safety claims on self-improving agent runtimes are almost always self-graded: a policy file, a guardrail, or a README commitment.
Runtime assurance (RTA) protects a safety-critical system by switching from an advanced controller to a verified safe controller when a monitored condition is violated. The standard latching rule, which trips on the first breach of the safe set and then coasts, is correct for a diverging controller but pathological for a capable online-adapting one.
arXiv:2607. 13070v2 Announce Type: replace-cross Abstract: Safety claims for self-improving agent runtimes are almost always self-graded: a policy file, a guardrail, a promise in a README.
arXiv:2607. 07405v1 Announce Type: new Abstract: Tool-using LLM agents can violate the very policies they are deployed to enforce while appearing to complete the task successfully.