Get the latest tech news
Are LLMs able to notice the “gorilla in the data”?
A comparison of LLMs’ ability to perform exploratory data analysis
I’ll help analyze this dataset using the analysis tool first to understand the data distribution and relationships, then create visualizations to illustrate the findings. The data points form distinct curves and lines across the plot, which is highly unusual for what should be natural, continuous biological measurements. The fact that both Sonnet and 4o required explicit prompting to notice even dramatic visual patterns suggests they may miss crucial insights during open-ended exploration.
Or read this on Hacker News