Quantum
My new course at UT Austin: AI Alignment Theory
Publication: Shtetl-Optimized
Author: Scott
Source published: 04/10/2026, 22:13:33 UTC
Briefing created: 04/10/2026, 23:18:42 UTC
Based on a feed excerpt; full article content has not been acquired. Source evidence retrieved 04/10/2026, 22:16:25 UTC.
Briefings
Professor Scott Aaronson at UT Austin teaches CS395T AI Alignment Theory, a discussion seminar focused on theoretical foundations rather than canonical theorems. The course began with Eliezer Yudkowsky's 2022 essay 'AGI Ruin: A List of Lethalities,' which outlines catastrophic risks of misaligned AGI systems. Students submitted written reports on these papers and class discussions, with the first report by Tennyson Bardwell detailing the essay's three sections on magnitude, prediction difficulties, and potential outcomes. The course is currently open to students who have shown interest, though the specific selection criteria remain unverified.
Key points
- Professor Scott Aaronson teaches CS395T AI Alignment Theory at UT Austin as a discussion seminar focusing on theoretical and mathematical foundations rather than canonical theorems.
- The course's first reading assignment was Eliezer Yudkowsky's 2022 essay 'AGI Ruin: A List of Lethalities,' which outlines catastrophic consequences of misaligned AGI systems and is divided into three sections on magnitude, prediction difficulties, and potential outcomes.
Context and limitations
The course is currently open to students who have shown interest, but specific selection criteria and enrollment numbers are not provided. The first report presented was by Tennyson Bardwell, and future reports are expected over the coming weeks, though their specific content and dates are not detailed.
Generated by Bittrees-hosted qwen35-2b sha256:aaf42c8b7c3c on 04/10/2026, 23:18:42 UTC. Based on publisher-provided feed evidence; may not cover the full original article. Check the source for details.
View archived briefing on IPFS
bafybeic4kdce7tyj5er7t47pgf6xumdpo4y5exoawqumjn2gbhzhw6qmte