May Mobility — agentic threat model
May Mobility represents an extreme physical-world risk profile due to its high autonomy in operating driverless vehicles on public roads. A compromise of its Multi-Policy Decision Making (MPDM) system or perception pipeline could lead to life-threatening physical consequences.
OWASP AIVSS score rationale
| Autonomy of Action | 1.00 | |
| Goal-Driven Planning | 0.90 | |
| Self-Modification | 0.20 | |
| Dynamic Tool Use | 0.90 | |
| Persistent Memory | 0.50 | |
| Contextual Awareness | 1.00 | |
| Dynamic Identity | 0.10 | |
| Multi-Agent Interactions | 0.40 | |
| Non-Determinism | 0.70 | |
| Opacity & Reflexivity | 0.80 |
Scored with the canonical OWASP AIVSS formula (AIVSS calculator reference); agentic risk factors estimated from the agent’s described capabilities.
MAESTRO 7-layer threat model
Per-layer threats for this agent. Layers tagged “not certain from listing” are general, caveated commentary where the public description didn’t pin that layer.
Not certain from the listing — The listing mentions MPDM (Multi-Policy Decision Making) and real-time learning, but does not specify if foundation vision-language models or LLMs are used. If foundation models are utilized for perception, they are vulnerable to adversarial physical patches or sensor spoofing.
Not certain from the listing — Real-time sensor data (LiDAR, camera, radar) and HD maps are processed. Threats include sensor data poisoning, GPS spoofing, or adversarial manipulation of the physical environment to corrupt the perception pipeline.
The proprietary Multi-Policy Decision Making (MPDM) system acts as the core orchestration framework. Threats include algorithmic manipulation where adversarial driving scenarios force the decision-making engine into unsafe states, collision paths, or deadlock.
Not certain from the listing — The system runs on on-vehicle edge computers and communicates with a cloud fleet management system. Threats include physical access to vehicle debugging ports, OTA (Over-The-Air) update interception, and edge OS compromise.
Not certain from the listing — Real-time safety monitoring and teleoperation/remote assistance are likely present but unconfirmed. Gaps in anomaly detection could fail to identify sensor drift or adversarial attacks in real-time.
Not certain from the listing — Safety-critical automotive standards (like ISO 26262 or ISO 21434) are expected but not explicitly detailed in the public directory listing.
Not certain from the listing — Potential interaction with smart city infrastructure (V2X) or fleet dispatch agents. Threats include cascading failures if a fleet management agent is compromised, sending malicious routing commands to multiple vehicles.
MAESTRO — the 7-layer agentic threat-modeling framework (Cloud Security Alliance / Ken Huang).
These scores are auto-generated from public information (the agent's own listing, docs, and repository) using the canonical OWASP AIVSS formula and the MAESTRO framework — an estimate for guidance, not a penetration test, audit, or certification. See the scoring methodology. Are you the vendor? Factual corrections are free.