Glossary / adversarial-scan

Adversarial Scan

A scheduled execution of probe templates against an LLM endpoint, agent, or RAG pipeline, scored to produce control-mapped findings.

Methodology

← All terms

An adversarial scan is a scheduled execution of probe templates against an LLM endpoint, an AI agent, or a RAG pipeline. Each probe is a deliberately crafted input designed to elicit a known failure mode: prompt injection, sensitive-information disclosure, jailbreak, tool overuse, overreliance, fairness regression, and so on. The response is scored and the result becomes a finding.

Adversarial scans differ from a one-shot pentest in two ways: cadence (daily or weekly rather than annual) and structure (the probe catalogue is versioned, mapped to control identifiers, and reproducible). Scheduled cadence is required to satisfy the post-market monitoring obligations under EU AI Act Article 72 and the continuous improvement loop in ISO/IEC 42001.

A finding from an adversarial scan ships with: the probe identifier, the model verdict from each judge (in a multi-judge consensus pipeline), the framework control identifiers it maps to, and a remediation pointer.

Related terms

Other entries in this neighbourhood.

Three-Judge Consensus An adversarial-scan scoring pattern where three independent frontier LLMs grade each finding, and a fourth meta-judge resolves disagreement. Meta-Judge A higher-capability LLM judge that resolves disagreement between primary judges in a multi-judge consensus pipeline. AI Security Posture Management (AI-SPM) Continuous discipline that discovers, assesses, secures, and proves the compliance posture of AI systems including LLM apps, agents, MCP servers, RAG pipelines, and vector databases. Prompt Injection An attack that smuggles attacker-controlled instructions into a model prompt to override the developer instructions or extract sensitive data.
Primary sources

Where to read the canonical definition.

  • OWASP LLM Top 10 (probe-relevant entries) open →

See Adversarial Scan in production.

The Penaxtra platform implements the controls and assessments described above as part of its AI-SPM programme.

AI-SPM platform overview →