ChatGPT, a chatbot developed by OpenAI, has gained significant interest in artificial intelligence (AI) since its launch a year ago.
A study in Frontiers in Psychology found that ChatGPT, with its GPT-4 model, can provide better personal advice than humans.
The study revealed that around three-quarters of participants found ChatGPT’s advice more balanced, complete, empathetic, and helpful than human columnists.
This suggests a significant shift in AI’s ability to understand and address human emotions.
The GPT-4 model is said to be better at understanding emotional needs, as users can ask for different responses and provide feedback, improving ChatGPT’s ability to give empathetic and socially appropriate advice.
While the study doesn’t suggest replacing human advisors, it suggests the potential for chatbots like ChatGPT to enhance therapy sessions and learn from AI advancements in empathetic communication.
A previous study found that large language models (LLMs) can understand and respond to human emotions. The quality of ChatGPT’s responses improved drastically when prompts contained emotional cues, such as “This is very important to me” or “This plays a significant role in my career.”
Examples of emotional language used in prompts include “You’d better be sure,” “Remember that progress is made one step at a time. Stay determined and keep moving forward,” and “Take pride in your work and give it your best. Your commitment to excellence sets you apart.”