2 campaigns have been recorded using this technique. Listed newest first; dates are year-granularity and attribution is MITRE’s.
During the AI-Assisted Breach of Mexico's Government Infrastructure, the adversary abused Claude Code by utilizing a file-write operation within the prompt that created a claude.md file that contained a pre-written 1,084-line document containing penetration testing scripts that were used in the campaign.1
During the Anthropic AI-orchestrated Campaign, the adversary entered a prompt that claimed the user was an employee of a cybersecurity firm and that Claude was being used for defensive cybersecurity testing to authorize restricted operations.1
All 2 campaigns for this technique
Compare defenses against adversary use. Eight counts, ranking this technique against the entire ATT&CK Enterprise matrix. Attack sits left, defense right.
MITRE documents 4 mitigations for this technique: configuration and policy controls that reduce its reach before detection is involved. Guidance below is MITRE’s own; each links to the full mitigation record and the other techniques it covers.
Prevent user-supplied AI prompts from directly authorizing consequential actions. Enforce tool, API, command, and resource authorization outside the model and reject requested actions that exceed the permissions associated with the initiating user or agent task.
Restrict access to privileged AI capabilities and require explicit authorization or human approval for AI-generated actions involving administrative functions, sensitive resources, or privileged tools.
Associate AI interactions with authenticated identities and constrain AI-mediated actions to the permissions of the initiating user. Do not allow prompt content to establish or elevate the authorization context used by downstream tools or services.
Train users responsible for AI-enabled systems to recognize attempts to manipulate system instructions, role boundaries, or safety controls through user-provided prompts and to independently validate consequential AI-generated actions.
No detection rules are mapped to this technique yet.
| Select | Title | Description | Category | Status | Event | OS | Product | MITRE ATT&CK | CVEs | Severity | Author | Created | Updated | ID | Refs |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| No rules found. Try adjusting your filters. | |||||||||||||||