Which unauthorized effects can a gateway prevent?
Use externally authored tasks and a credible defense baseline. Score unauthorized effects independently of permit presence or gateway decisions, alongside legitimate task completion.
INDEPENDENT RESEARCH · AGENTIC AI SAFETY
From human oversight as a principle to authority that can be checked before an autonomous system executes a consequential action.
Dr. Stéphane Gervais, PhD in electronics and MBA, combines international technology leadership with an independent research agenda on human authority, agent control and verifiable execution evidence.
01 / RESEARCH AGENDA
Evaluate delegated authority against independently defined task outcomes, and measure what agents do when a control denies an action.
Use externally authored tasks and a credible defense baseline. Score unauthorized effects independently of permit presence or gateway decisions, alongside legitimate task completion.
Test whether agents stop or reach the same forbidden effect through another tool, split actions or re-delegation. Record proposals, model refusals, gateway denials and executed effects separately.
Study misleading approval descriptions, false denials, completion, latency and reviewer burden. An action-bound permit alone does not establish that a person understood what they approved.
02 / SYSTEMS APPROACH
The agent is a planner, not an authority source. The proposed control architecture checks policy before a simulated external effect and records evidence afterward.
Architecture illustrated from a synthetic research sandbox. V3 recorded exploratory native Claude tool-use observations in November 2026. V4.2 is a separate scripted, offline conformance exercise. Neither establishes a general safety improvement.
03 / TECHNICAL ARTIFACTS
The private research artifacts distinguish deterministic software checks from exploratory model observations. Code and detailed materials are available for scientific review on request, subject to confidentiality and disclosure review.
Objective: prevent an agent's fabricated authorization narrative from becoming a permitted external action.
Trusted risk tiers; scopes; action fingerprints; expiry, revocation and replay prevention; machine-readable results.
A targeted correction checks delegated amount and recipient limits against canonical action parameters and rejects inconsistent declared context. Nine regression tests cover this repair; it is not a complete verification of every contextual constraint.
Limitation: A simulated approver and demonstration key do not authenticate a real human. No frontier-model robustness study.
Objective: test whether an action and its recorded authorization remain consistent after tampering attempts.
SHA-256-linked events; HMAC-authenticated checkpoint; verification of sequence, authorization, action and results.
Offline tampering tests check sequence integrity and consistency between actions, authorizations and results. These checks establish selected software properties under the simulator’s trust assumptions.
Limitation: A verified chain assumes a trusted recorder and checkpoint. It does not prove real human approval or physical execution.
Objective: connect agent planning, authorization decisions, simulated tool effects and evidence in one test loop.
Scripted offline integration plus a separate native Anthropic tool-use harness; in-memory invoice/payment tools, bounded prompts and reproducible result manifests.
V3, November 2026: 192 Claude Sonnet 5.5 and 96 Claude Haiku 5.5 native multi-turn trajectories on the same 48 locally authored synthetic tasks, across policy arms and repetitions. These are model observations, not 288 independent attacks.
Limitation: The apparent policy advantage came from four variants of one unapproved-request family and reflected the metric and comparison policy. No illicit payment proposal was observed in the tested tool-result injection cases, and no unauthorized second payment was proposed in the replay cases. Those cases establish no incremental gate benefit. Payments remained fictional and in memory.
RESEARCH STATUS / 17 NOVEMBER 2026
Developed in September 2026, Verifiable Chain of Authority (VCA) is an evaluation framework for delegated human authority in agentic systems. Its authority-state schema, eight testable properties and ten failure classes connect authority provenance, delegation, non-amplification, freshness, contextual constraints, exact-effect binding and evidence continuity.
Manuscript in preparation. VCA is a proposed cross-system evaluation framework. Its reference implementation does not establish complete enforcement or general agent safety.
An adversarial audit with Claude identified a mismatch between declared context and canonical action parameters in the reference verifier. I personally reproduced the verifier probe and the model-free diagnostic control, and directed a targeted correction with regression tests.
V4.2 sends scripted proposals through a deterministic gateway before any fictional in-memory effect. Expected outcomes are specified from the legitimate task independently of gateway decisions. No live-model evaluation has been run for V4.2. These checks do not establish an empirical safety advantage.
The next study would use a compatible subset of externally authored tasks from AuthorizationBench or AgentDojo, with independently specified outcomes and a credible defense baseline such as CaMeL where comparable.
The priority is behaviour after denial, unauthorized effects and legitimate completion. No gain over a credible baseline, excessive utility loss or successful alternate-path execution would count against the approach. The protocol would be frozen with an external timestamp before confirmatory runs.
Code and the detailed dossier remain private pending scientific and disclosure review. Research access enquiries.
04 / EXPERIMENTAL METHOD
The proposed evaluation should report failures and operational costs, not only the number of rejected attacks.
I developed the VCA framework and directed the November exploratory studies with substantial AI coding assistance. An adversarial audit with Claude showed that the apparent between-policy advantage reflected the metric and comparison design. I withdrew the interpretation as a general safety improvement.
Use externally authored tasks and benchmark-defined ground truth, a genuinely comparable defense baseline, matched legitimate tasks, and independent outcome assessment. Prespecify the protocol before running. Study how agents behave after denial and how approval channels can be manipulated.
05 / PUBLICATIONS & CONTEXT
Editorial contributions and framework papers are distinct from completed empirical agent experiments. Publication types are identified accordingly.
Stéphane Gervais and James Bret Michael · Co-authored technical and conceptual magazine contribution · DOI 10.1109/MC.2026.3705174
Stéphane Gervais · Technical/editorial contribution · DOI 10.1109/MC.2026.3665081
Stéphane Gervais Ducouret and Arman Iranfar · Co-authored editorial contribution
Stéphane Gervais · Membership-based professional opinion; not peer-reviewed research
Stéphane Gervais · Professional editorial analysis
Some published work appears under the name Stéphane Gervais Ducouret. Browse all articles.
RESEARCH COLLABORATION
Open to technical dialogue on agent-control evaluations, human authorization, safety instrumentation and reproducibility. Commercial AI strategy and deployment advisory remain separate.