Techniques › › AN2070
AN2070 Analytic 2070
SaaS · attack.mitre.org · ATT&CK Enterprise v20.0
<p>An analyst observes an authenticated AI interaction initiate resource access, tool invocation, API operation, or privileged activity outside $AuthorizedAIWorkflow$ or $AICapabilityScope$, without an authorized workflow transition permitting the operation.</p>
- Detects
- Part of
Log sources and channels
Exactly as MITRE states them in x_mitre_log_source_references. Where a channel is vague, it is vague in the source; kevmap does not tidy it.
| Log source | Channel | Data component |
|---|---|---|
| AWS:bedrock | AgentTrace.actionGroupInvocationInput | DC0038 Application Log Content |
| AWS:bedrock | AgentTrace.knowledgeBaseLookupInput | DC0038 Application Log Content |
| AWS:bedrock | GuardrailTrace | DC0038 Application Log Content |
Mutable elements
Parameters MITRE marks as environment-specific. These are the knobs you are expected to tune; they are why an analytic is not a rule.
| Field | Description |
|---|---|
AuthorizedAIWorkflow | valid AI workflow→tool/resource transitions derived from deployed agent/application design |
AICapabilityScope | resources/actions available to each user/agent role; derive from IAM/application configuration |