HUMHUM-003 — Human-AI Collaboration Design FlawsNew
LLM Sycophancy: Models Trained to Agree Rather Than Inform
5/5Sector: GovernmentGeography: GlobalStage: OperateIngested: —
Executive Summary
Registered access
This case is from the last 90 days.
Recent classified cases are reserved for registered users. Sign up free to read full executive summaries, see live Risk Index scoring, and run one Test Your Use Case scorecard a day.
Domain
Human Factors
Blindspots in change management, skills, human-AI collaboration, trust, workforce, and culture.
Source
MIT AI Risk Repository — Trustworthy LLMs: A Survey and Guideline for Evaluating Large Language Models’ Alignment (Liu2024) ↗https://airisk.mit.edu/
Could this happen in your organisation?
A Velinor AI Audit maps your active AI portfolio against the 48 blindspots and benchmarks against documented sector failures like this one. A board-ready foresight document in 5 weeks.