“Implications of the inference scaling paradigm for AI safety” by Ryan Kidd

Scaling inference With the release of OpenAI's o1 and o3 models, it seems likely that we are now contending with a new scaling paradigm: spending more compute on model inference at run-time reliably improves model performance. As shown below, o1's AIME accuracy increases at a constant rate with the logarithm of test-time compute (OpenAI, 2024). OpenAI's o3 model continues this trend with record-breaking performance, scoring: 2727 on Codeforces, which makes it the 175th best competitive programmer on Earth; 25% on FrontierMath, where "each problem demands hours of work from expert mathematicians"; 88% on GPQA, where 70% represents PhD-level science knowledge; 88% on ARC-AGI, where the average Mechanical Turk human worker scores 75% on hard visual reasoning problems. According to OpenAI, the bulk of model performance improvement in the o-series of models comes from increasing the length of chain-of-thought (and possibly further techniques like "tree-of-thought") and improving the chain-of-thought [...] ---Outline:(02:52) AI safety implications(04:04) AGI timelines(04:55) Deployment overhang(06:12) Chain-of-thought oversight(07:45) AI security(09:10) Interpretability(10:07) More RL?(11:31) Export controls(11:57) ConclusionThe original text contained 4 footnotes which were omitted from this narration. The original text contained 7 images which were described by AI. --- First published: January 15th, 2025 Source: https://forum.effectivealtruism.org/posts/bYXjejHrvq65jFL9s/implications-of-the-inference-scaling-paradigm-for-ai-safety --- 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.