Podcast 2
Published:
Details about this podcast: In this episode we explore how different large language models respond to the same prompt and what that reveals about their accuracy, reliability, and ethics. We talk about how we designed our prompt, why we chose certain models, and what kinds of outputs we received including examples of when the models got things wrong. We then compare their performances, reflect on the risks of relying on AI for information, and explain why hallucinations matter for research, journalism, and public trust. We end by sharing what we learned and giving advice for anyone planning to use language models for academic or professional work.
Listen:
Watch: