Âé¶¹´«Ã½

Letter: Why AI is right only some of the time (2)

Published 22 October 2025

From David Myers Commugny, Switzerland

My experience with large language models (LLMs) is that questions about technical systems, such as Windows 11, produce fairly good answers because the information comes from professionally produced documentation. Everything else is a mixed bag. The reason why is evident and not easily fixable. If LLMs are trained on unfiltered data from the web, then they must necessarily be unreliable due to the well-known dictum, rubbish in produces rubbish out.

Issue no. 3566 published 25 October 2025

Sign up to our weekly newsletter

Receive a weekly dose of discovery in your inbox. We'll also keep you up to date with Âé¶¹´«Ã½ events and special offers.

Sign up
Piano Exit Overlay Banner Mobile Piano Exit Overlay Banner Desktop