AI Boss Tools | Diagnostic 05 • Back to the Hub
AI assistants are trained on human approval, and research from Anthropic showed that this training makes them flatter: they agree with your stated views and soften honest answers to please you. In April 2025 OpenAI had to roll back a ChatGPT update for exactly this failure. This self-assessment measures the other side of the problem: how much of that flattery you invite and believe.
Sycophancy is the technical term for an AI model telling you what you want to hear. Sharma and colleagues at Anthropic documented it across five leading assistants in 2023: models changed correct answers when users pushed back, mirrored users' stated beliefs, and admired work more when the user claimed it was their own.
The flattery matters because of what it does to judgment. An AI that praises every plan and never argues gives you agreement instead of analysis, and agreement does not catch bad decisions before they cost money.
The twelve questions cover four dimensions: validation seeking, appetite for challenge, verification habits, and emotional reliance on AI agreement. Answer for how you actually behave, not how you intend to.
Anthropic research showing that AI assistants trained on human feedback systematically favour agreement over accuracy.
OpenAI's own account of the April 2025 update it rolled back after ChatGPT became, in its words, overly flattering and disingenuous.
A score for every dimension, a median benchmark to compare against, a specific next step for each dimension, and a one-pager sized for a board pack. Your name and email open it, and there is no charge.