Atlas AI — agentic threat model
Cognite Atlas AI presents a high-risk profile due to its integration with critical industrial infrastructure (Energy & Utilities) and its capability to automate complex workflows, where non-deterministic AI decisions could lead to physical or operational disruptions.
OWASP AIVSS score rationale
| Autonomy of Action | 0.80 | |
| Goal-Driven Planning | 0.70 | |
| Self-Modification | 0.20 | |
| Dynamic Tool Use | 0.80 | |
| Persistent Memory | 0.60 | |
| Contextual Awareness | 0.90 | |
| Dynamic Identity | 0.40 | |
| Multi-Agent Interactions | 0.50 | |
| Non-Determinism | 0.70 | |
| Opacity & Reflexivity | 0.60 |
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 — autoLLM capabilities suggest dynamic model selection or abstraction, but specific foundation models, alignment guarantees, or protections against adversarial prompt injection are not disclosed.
Integrates directly with Cognite Data Fusion® and an Industrial Knowledge Graph. Threats include knowledge-base poisoning of industrial data, which could lead to unsafe physical operations, and data exfiltration of sensitive utility/energy schematics.
The low-code agent builder orchestrates workflow automation and decision-making. Risks include insecure tool integration and tool misuse, where automated workflows could execute unintended physical or operational commands in industrial systems.
Not certain from the listing — deployment details are abstracted under 'easy deployment' and Cognite Data Fusion® integration, but lack specific sandboxing, network isolation, or container security details for the executed agents.
Not certain from the listing — no explicit mention of evaluation, logging, or real-time guardrails for the generated industrial agents is provided in the directory listing.
Not certain from the listing — while it extends Cognite Data Fusion® which has enterprise-grade security, specific identity, authorization, and compliance controls for Atlas AI are not detailed.
Not certain from the listing — supports 'out-of-the-box industrial agents' but does not explicitly detail multi-agent collaboration, marketplace trust, or cascading failure protections.
MAESTRO — the 7-layer agentic threat-modeling framework (Cloud Security Alliance / Ken Huang).