Ep. 35 - Doing research with AI-generated users

In this episode, we delve into the intriguing world of AI and explore how AI-generated users can be integrated into the research process. I share my experiences and experiments with various local versions of large language models (LLMs), such as GPT, Mistral, and Llama, and discuss the important distinctions between human and synthetic user feedback. We also touch on the potential benefits and limitations of using synthetic users, emphasizing that while they have their place, they can never replace real human interactions.(0:11) Introduction to AI Series(2:21) Exploring AI-Users(4:48) Comparing AI-gen users and Real User Feedback(8:50) Challenges and Limitations of AI-gen Users(12:38) Practical Applications and Training with AI-gen Users(16:09) Conclusion and Future EpisodesIMPORTANT: During the episode, I use terms like AI-users and synthetic users interchangeably. In every case, I’m generically referring to AI-generated users and not specifically to the brand “Synthetic Users”.