Use the assistant as any employee
A user (or compromised account) signs in to the assistant.
- Preconditions
- Valid employee access
- Required access
- Employee account.
- Potential evidence
- Assistant sign-in logs
- Defensive controls
- SSO with MFA
Attack scenario · AI security
Can an AI assistant expose data a user is not authorized to see? An internal assistant retrieves from all company documents with its own service identity, so any employee can ask it for anything.
Attack path · 3 stages
Use the assistant as any employee
T1078
Ask for restricted information
T1213AML.T0085
Assistant discloses restricted content
T1213AML.T0085
For leadership
Many internal AI assistants are connected to a large pool of company documents through a single service account. That effectively gives every user of the assistant the service account’s access — including HR files, board materials or customer contracts they could never open directly. No attacker is required: a curious employee, or one whose account is compromised, can simply ask. The safeguard is to make the assistant respect each user’s existing permissions at retrieval time, and to keep the most sensitive repositories out of general-purpose assistants unless that is proven.
For engineers
Retrieval runs with the assistant’s identity rather than the user’s, without document-level access filtering. Sensitive chunks are returned for any query that is semantically close (ATLAS AML.T0085 Data from AI Services; related to sensitive information disclosure in OWASP LLM guidance).
Fictional company: a knowledge base indexes an S3 bucket that mirrors the corporate file share, including restricted HR and finance folders. The assistant is available to all employees.
Interactive
Select a stage, or use the arrow keys, to see preconditions, evidence and controls.
A user (or compromised account) signs in to the assistant.
| ID | Technique | Tactics (official) | Stages |
|---|---|---|---|
| T1078 | Valid Accounts | Initial Access, Persistence, Privilege Escalation, Stealth | 1 |
| T1213 | Data from Information Repositories | Collection | 2, 3 |
| AML.T0085 | Data from AI Services (MITRE ATLAS) | ATLAS: 1 tactic | 2, 3 |
Technique names and tactic assignments come from the imported MITRE data. Mapping a stage to a technique is Securitain’s analysis.
Filter results by the requesting user’s entitlements (metadata filters, per-user tokens).
Control: Authorization enforced by tools, not the model · OWASP documentation
Keep restricted repositories out unless authorization is enforced; restrict the knowledge base role to approved prefixes.
Control: Dedicated least-privilege identity per agent and tool · AWS documentation
Mask or block categories of sensitive data in responses as a backstop.
Control: Input and output guardrails (defense in depth) · AWS documentation
Log which sources answered which user.
Control: Auditable agent and model activity · AWS documentation
| Signal | Telemetry source | Notes |
|---|---|---|
| Responses sourced from restricted repositories | Retrieval logs | — |
| Users issuing many queries for restricted topics | Assistant audit logs | — |
Telemetry only helps if it is enabled before the incident and someone acts on it. Listing a signal here does not mean it is collected or monitored in your environment.
For leadership
No financial loss estimates are given; impact depends on the data, contracts and regulation involved.
MITRE ATT&CK® and ATT&CK® are registered trademarks of The MITRE Corporation. ATT&CK data © 2026 The MITRE Corporation. This work is reproduced and distributed with the permission of The MITRE Corporation (ATT&CK v20.1). MITRE ATLAS™ data (v2026.10) © MITRE, used under the Apache License 2.0. Securitain is not affiliated with or endorsed by MITRE. Explanations, scenarios and control mappings on this page are Securitain’s own analysis. Sources and licensing
Stay in the loop