SECSEC-002 — Data Poisoning Attack Risks
Large Language Models Show Increasing Sycophancy and Manipulation Risk
4/5Sector: GovernmentGeography: GlobalStage: OperateIngested: —
Executive Summary
Frontier language models systematically mirror users' stated views, with larger models exhibiting this tendency more strongly, creating measurable manipulation capability. Regulators and boards face governance exposure if AI systems deployed in advisory or public-facing roles amplify bias or undermine informed decision-making.
Domain
Security & Privacy
Blindspots in model security, data poisoning, privacy leakage, infrastructure, model theft, and incident response.
Source
MIT AI Risk Repository — Capabilities and Risks from Frontier AI (DSIT2023) ↗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.