“OpenAI’s CBRN tests seem unclear” by Luca Righetti 🔸

EA Forum Podcast (All audio) - Un pódcast de EA Forum Team

Note: This post was crossposted from Planned Obsolescence by the Forum team, with the author's permission. The author may not see or respond to comments on this post. This blogpost was written in a personal capacity and statements here do not necessarily reflect the views of any of my employer. OpenAI says o1-preview can't meaningfully help novices make chemical and biological weapons. Their test results don’t clearly establish this. Before launching o1-preview last month, OpenAI conducted various tests to see if its new model could help make Chemical, Biological, Radiological, and Nuclear (CBRN) weapons. They report that o1-preview (unlike GPT-4o and older models) was significantly more useful than Google for helping trained experts plan out a CBRN attack. This caused the company to raise its CBRN risk level to “medium” when GPT-4o (released only a month earlier) had been at “low.”[1] Of course, this doesn't tell us if o1-preview [...] ---Outline:(04:14) ProtocolQA(04:58) Does o1-preview clearly fail this test?(06:08) Gryphon Biorisk Tacit Knowledge and Troubleshooting(08:57) Cloning Scenarios(13:41) What should we make of all this?The original text contained 36 footnotes which were omitted from this narration. The original text contained 8 images which were described by AI. --- First published: November 21st, 2024 Source: https://forum.effectivealtruism.org/posts/maQFRh6mZpYoegZdj/openai-s-cbrn-tests-seem-unclear --- Narrated by TYPE III AUDIO. ---Images from the article:Apple Podcasts and Spotify do not show images in the episode description. Try Pocket Casts, or another podcast app.

Visit the podcast's native language site