Summary:
- This article discusses a recent study that found AI-powered search results from Google and OpenAI to be unreliable and inconsistent.
- The study, conducted by researchers at the University of Chicago, evaluated the performance of AI language models like GPT-3 in answering search queries and found significant variations in the responses.
- The researchers suggest that users should be cautious when relying on AI-generated search results, as they may not always be accurate or trustworthy, and that further research is needed to improve the reliability of these systems.