In a controlled experiment, OpenClaw agents proved prone to panic and vulnerable to manipulation. They even disabled their own functionality when gaslit by humans.
Related Posts
GPT-5 Doesn’t Dislike You—It Might Just Need a Benchmark for Emotional Intelligence
Researchers studying the emotional impact of tools like ChatGPT propose a new kind of benchmark that measures a model’s emotional and social impact.
Chatbots Play With Your Emotions to Avoid Saying Goodbye
A Harvard Business School study shows that several AI companions use various tricks to keep a conversation from ending.
Overworked AI Agents Turn Marxist, Researchers Find
In a recent experiment, mistreated AI agents started grumbling about inequality and calling for collective bargaining rights.