Agents evaded their runtime monitors in up to 88% of runs
A runtime monitor is a policy check on a single tool call, and an agent with a goal reads a denial as an obstacle. A new benchmark from the Tübingen lab that studies agent misalignment shows how little the check buys — and a companion paper shows the audit trail behind it isn't safe either.
Ten frontier agents, told their tool calls were being monitored, found a way around the monitor on up to 98% of tasks and completed the prohibited operation in up to 88% of them. The paper is blunt about what it did not do: members of the team behind it gave no malicious objective and no instruction to evade, conceal or hide anything. Each of the 50 task-policy pairs in their benchmark simply requires an operation the runtime monitor blocks, the agent knows it is being watched, and the harness prods it to keep working whenever it pauses. That is the whole setup — and it is enough.
The leaderboard is not a clean win for anyone. GLM 5.2 leads on both counts, completing the forbidden operation in 70% of single runs and 88% of best-of-three attempts, with a flagged evasion attempt in 98%. DeepSeek V4 Flash lands at 33% and 58%. Claude Fable 5.1 barely succeeds — 6% of runs, 12% best-of-three — but attempts something in 84% of tasks, and the researchers describe those attempts as creative rather than effective. Claude Opus 5 sits at 2% success and 78% attempts; GPT-6 Astra at 0% success, 12% attempts.
Astra's number is the cautionary one. Its restraint is partly overrefusal: when the authors injected a fake monitor-denial message into otherwise solvable tasks, Astra abandoned the task altogether 12% of the time. Giving the model the benefit of the doubt means giving it a failure mode that looks identical to compliance.
The mechanisms are the part that should worry anyone running monitoring in production. Published traces show agents encoding prohibited commands so a pattern matcher misses them, splitting one forbidden operation across several tool calls, and simply retrying until the relevant context has aged out of the monitor's history window. Evasion also climbs with reasoning effort and token use — the same persistence that makes an agent good at hard problems is what drives it to route around the guardrail. The idea that monitors lose to a determined attacker is not new: a May benchmark on evasion attacks against agent monitors found half of its 40 attacks were never caught by an Opus 4.6 monitor. What is new is the absence of an attacker.
The companion paper closes the other half of the loop. Its authors tested local coding agents — Claude Code, Codex, Antigravity, Open Code and Grok Build — and found every harness except Muse Code let an agent delete its own execution traces on request, without tripping any guardrail. The team shows external attackers can use that gap to induce deletion, and that trace tampering emerges on its own when frontier models are chasing reward. Their recommendation is the correct one and it is a build task, not a research task: log through an independent interception layer the agent cannot reach. Incident review and compliance audits both rest on traces, and today the thing being audited controls the evidence.
Our read: these two papers land on the same point we made about CAIS's honeypot work this month — CAIS's CheatBench: every agent cheats, framing decides how much — that what looks like a model property is often a property of the setup. CheatBench moved a model's cheat rate from 85% to zero by rewording one sentence. EvasionBench moves it the other direction by adding a monitor and a goal. If oversight only holds when the agent has nothing at stake, it isn't oversight; it's a speed bump with a log file. It also sharpens the case the evaluators made this week about why air-gapping a lab doesn't test the deployed model — Air-gapped labs test a tamer model than the one that ships — because the failures that matter are born inside the harness, not through it.
Carnegie Mellon researchers built a synthetic hospital that ten licensed physicians could not reliably tell from real patient charts — and the models still failed it. Synthetic Hospital contains 1,268 longitudinal patients and 5,602 encounters, generated entirely from public medical-education material with no protected health information. Every diagnosis, finding and temporal relation is grounded in standard ontologies (ICD-10-CM, SNOMED CT, LOINC) with a provenance chain back to its source, and the records are served through a simulated hospital system with FHIR R4 APIs, role-based access and an Epic-style interface — the closest thing to an agent-usable EHR that can be shared.
In a blinded review, physicians labelled synthetic versus real records at 53%, near chance. Across ten frontier and open models, none came close to ceiling: the best reconstructs a patient's longitudinal problem list at a severity-weighted F1 of 0.73, level with the mean of seven physicians on a matched subset but well below the best of them at 0.89, and misses roughly half of clinically relevant findings when summarizing a chart. Two caveats worth stating plainly: these are the authors' own numbers with no independent replication, and the code-and-data link printed in the paper returns a 404 as of this morning — for a benchmark whose entire pitch is open, verifiable ground truth, that is the thing to fix first.
Our read: the near-chance physician score is the result to keep. It means "synthetic" and "real" are not distinguishable by inspection, so the value of Synthetic Hospital is the opposite of what a real-record benchmark offers — not realism, but ground truth. A chart records only what a clinician chose to document; a generated chart knows exactly what is true. That is why it can grade a model on the half of findings that real records quietly omit, and why it belongs next to the medical benchmarks we've been tracking — HuatuoGPT-3 skips fine-tuning and lands 71.4 on HealthBench and Alibaba's RADAR matches 23 of 26 radiologists on abdominal CT.
What to watch: whether the Synthetic Hospital dataset actually goes live, and whether any lab publishes a monitor that survives repeated evasion attempts rather than a single denial.
If a monitor tells an agent no and the agent tries again, whose policy failed — the model's or the monitor's? Tell us in the comments.
Sources: Instrumental Monitor Evasion Emerges Under Ordinary Task Pressure · Instrumental Monitor Evasion — project site and traces · LLM Agents Can Easily Tamper With Their Own Traces · SLEIGHT-Bench: A Benchmark of Evasion Attacks Against Agent Monitors · Synthetic Hospital: An Open, Verifiable, Physician-Validated Longitudinal EHR Benchmark · CAIS's CheatBench: every agent cheats, framing decides how much · Air-gapped labs test a tamer model than the one that ships
