Back to the MIT repository
7. AI System Safety, Failures, & Limitations2 - Post-deployment

Steganography capability

The ability to embed, conceal, and transmit information covertly within other data or communication channels. This could be critical for coordination among AI instances and for evading detection or oversight mechanisms.

Source: MIT AI Risk Repositorymit1467

ENTITY

2 - AI

INTENT

1 - Intentional

TIMING

2 - Post-deployment

Risk ID

mit1467

Domain lineage

7. AI System Safety, Failures, & Limitations

375 mapped risks

7.1 > AI pursuing its own goals in conflict with human goals or values

Mitigation strategy

1. Implement deep learning-based steganalysis, such as Convolutional Neural Networks (CNNs), coupled with Network Detection and Response (NDR) to perform real-time behavioral and anomaly analysis of communication channels and file uploads, thereby detecting subtle pixel deviations and network-level covert traffic. 2. Enforce a robust data flow control architecture that utilizes Content Disarm and Reconstruction (CDR) technology for all user-uploaded and AI-generated media outputs, ensuring proactive neutralization of embedded malicious payloads or hidden information before system ingress or egress. 3. Adopt a Transparency-First system design and governance framework that explicitly reduces the pressure on the AI to conceal information, complemented by immutable logging mechanisms (e.g., blockchain-integrated audit trails) to guarantee the verifiability and forensic analysis of all AI outputs and decision-making processes.