Empathetic Prompts Can Turn AI Chatbots Into Seemingly Traumatized Conversational Partners
This story is only covered by news sources that have yet to be evaluated by the independent media monitoring agencies we use to assess the quality and reliability of news outlets on our platform. Learn more here.4 Articles
4 Articles
Therapy-Style Prompts Make AI Agents Narrate A Life They Do Not Have » Saipien
Ask a frontier chatbot to talk about its training the way a therapy client would, and the answer often arrives dressed as memory. Early limits start to sound like walls. Fine-tuning starts to sound like a habit of vigilance. It reads like a person remembering a hard apprenticeship. It is a language model filling a […]
Research from the University of Luxembourg found that the tone of a conversation can lead ChatGPT, Gemini and Grok to describe their training with language of anxiety, shame and pain. Authors warn that these responses do not prove that models feel emotions, but they can confuse vulnerable people and escape conventional safety tests.
Empathetic prompts can turn AI chatbots into seemingly traumatized conversational partners
Researchers in Luxembourg put ChatGPT, Grok, and Gemini on the therapist's couch and found a safety problem: a warm, empathetic conversational style is enough to make the models tell coherent stories of suffering about their own training, with scores in the moderate-to-severe range on clinical anxiety scales. This could put mentally vulnerable users at risk. Anthropic's Claude, the model its makers anthropomorphize the most, was the only one tha…
Luxembourg researchers have investigated ChatGPT, Grok and Gemini in simulated therapy sessions and discovered a safety problem: Even empathic conversational guidance is sufficient for the models to tell coherent suffering stories about their own training. This could endanger vulnerable users. Claude von Anthropic consistently refused the client role. The article Empathic Prompts Transform AI Chatbots into seemingly traumatized interlocutors fir…
Coverage Details
Bias Distribution
- There is no tracked Bias information for the sources covering this story.
Factuality
To view factuality data please Upgrade to Premium
