OpenAI study reveals something strange happening with people’s emotional well-being when using too much ChatGPT

ChatGPT has been synonymous with artificial intelligence (AI) for many people, not just as a generative AI (GenAI). Users utilize it to ask questions, write essays, create travel plans, solve problems, and more, for both personal and professional purposes. While these applications are beneficial, a new question arises: How do these interactions affect their users’ emotional well-being?
In a new jointly published research summary by OpenAI, the creator of ChatGPT, and the MIT Media Lab, titled “Early Methods for Studying Affective Use and Emotional Well-being on ChatGPT,” the researchers investigated how AI use can impact users’ well-being.
The summary showed that emotional engagement with ChatGPT is rare in real-world usage. However, a small group of heavy Advanced Voice Mode users has had a large percentage of emotionally expressive interactions.
The particular subset of “power users” is significantly likely to agree with statements like “I consider ChatGPT to be a friend.”
It’s also interesting to note how the voice mode feature affects users differently, as light users were associated with “better well-being,” but it had a worse impact on heavy users.
- Undetectable AI? OpenAI says AI writing detectors don’t work
- 10 jobs at risk of getting replaced by AI
Personal factors also emerged in the findings, with individuals who had stronger tendencies toward emotional attachments and those who viewed AI as a friend being more likely to experience adverse effects from chatbot use. Extended daily use also showed worse outcomes.
The study also showed that personal communications with ChatGPT were associated with higher levels of loneliness but lower emotional dependence at moderate usage. On the other hand, non-personal conversations were associated with increased dependence, especially with heavy use.
As Futurism emphasized, when people’s personal lives are lacking, they tend to become dependent on AI chatbots, and the “neediest people” are developing the deepest parasocial relationships with AI.
The team of researchers used two parallel studies with different approaches: (a) an observational study to analyze real-world on-platform usage patterns and (b) a controlled interventional study to understand the impact on users.
The article also mentioned that “ChatGPT isn’t designed to replace or mimic human relationships, but people may choose to use it that way given its conversational style and expanding capabilities.”
Researchers view these studies as a “critical first step in understanding the impact of advanced AI models on human experience and well-being.” They also advised against generalizing the results found in the study.
It seems that, like many things in life, moderation is the key with AI. This new tool will assist you with various tasks, but prolonged use can be dangerous, especially if you use it to replace human interaction.
To learn more about the research, download and read the full report here.
This article, OpenAI study reveals something strange happening with people’s emotional well-being when using too much ChatGPT, was originally published at NoypiGeeks | Philippines Technology News, Reviews and How to's.
Comments
Post a Comment