Discussion: A Viable Strategy Exists for AI Value Loading
jd_pressman · x · 2026-07-19
Users discussed the classic AI safety problem proposed by Bostrom in 2014: how to inject high-quality human values into a model's objective function before it becomes too intelligent to control. The author believes that, despite some caveats, the academic community actually has a reasonable strategy to address the "value loading" problem. This strategy can handle the vast majority of standard inputs, and the rest can be managed as engineering issues.
Related event: Debate Rekindles Over AI Value Loading and Old Doom Models(4 posts)→
More from AGI Musings
- Claude Code skill uses 10 Markdown rules to make outputs ADHD-friendly — alex_verem · 2026-07-22
- AI Power Demand Exposes US Energy Gap, Urging Shift from Scarcity to Abundance — bradneuberg · 2026-07-22
- ControlAI CEO says an international ban on superintelligence is needed to avert extinction risk — zetalyrae · 2026-07-22
- Gary Marcus says LLMs still cannot really do math on their own — GaryMarcus · 2026-07-22
- Gary Marcus says LLM math skills are like knowing only a car’s engine size — GaryMarcus · 2026-07-22
- AI may make digital work infinitely leveraged while offline life gets more human — illscience · 2026-07-22