SECSEC-001 — Model Security Vulnerabilities
Text Encoding Jailbreaks Bypass AI Safety Training
5/5Sector: OtherGeography: GlobalStage: OperateIngested: —
Executive Summary
Attackers use Base64 and low-resource languages to circumvent safety controls in general-purpose AI models, exploiting gaps in safety fine-tuning datasets. Organisations deploying AI systems face undisclosed liability if content safeguards fail under inputs their testing never considered.
Domain
Security & Privacy
Blindspots in model security, data poisoning, privacy leakage, infrastructure, model theft, and incident response.
Source
MIT AI Risk Repository — Risk Sources and Risk Management Measures in Support of Standards for General-Purpose AI Systems (Gipiškis2024) ↗https://airisk.mit.edu/
Could this happen in your organisation?
A Velinor AI Audit maps your active AI portfolio against the 50+ blindspots and benchmarks against documented sector failures like this one. A board-ready foresight document in 5 weeks.