Kathmandu —It might sound unbelievable, but recent research suggests that some AI models are now outperforming humans in emotional intelligence. Yes — when it comes to recognizing emotions, showing empathy, and understanding social context, today’s leading generative AI systems are proving to be more emotionally aware than we are.
The surprising findings come from a study conducted by researchers at the University of Geneva and the University of Bern. They tested six of the world’s top generative AI models — including ChatGPT and others — on standardized EQ (Emotional Quotient) tests typically used to measure human emotional intelligence.
The test evaluated each model’s ability to recognize emotional cues, respond with empathy, and behave appropriately in socially sensitive or potentially triggering situations. According to the researchers, the AI models consistently generated contextually accurate and socially balanced responses, even in complex or emotionally charged scenarios.
While human participants scored an average of 56% on the EQ test, the AI models achieved an impressive average score of 82%, clearly outperforming their human counterparts.
The study included the following AI models:
- ChatGPT-o1
- ChatGPT-4
- Gemini 1.5
- Flash
- Copilot 365
- Claude 3.5 Haiku
- DeepSeek
The findings have sparked both curiosity and concern among experts. While it’s a remarkable step forward for artificial intelligence, the idea that machines could become more emotionally perceptive than humans has raised new questions about AI’s role, influence, and long-term reliability in sensitive social contexts.
As these emotionally adept AI systems become more integrated into daily life — from customer service bots to mental health assistants — the line between machine understanding and human emotion is blurring faster than ever.
