Bounded Regret
bounded-regret.ghost.io · 160 words · saved by 1 readers
AI, science, forecasting, philosophy
Cross-posted from the Transluce blog. To oversee an AI model, we'd ideally like to ask questions such as: * What are important situations where the model sandbags? * Does the model have Technically skilled people who care about AI going well often ask me: how should I spend my time if I think AI governance is important? By governance, I mean the constraints, incentives, and Currently, we primarily oversee AI with human supervision and human-run experiments, possibly augmented by off-the-shelf AI assistants like ChatGPT or Claude. At training time, we run RLHF, where humans (and/ This is…
saved by
related reading
- Foundation Models for Oversight | Transluce AItransluce.org
- Oversight Assistants: Turning Compute into Understandingbounded-regret.ghost.io
- Translucetransluce.org
- The case for ensuring that powerful AIs are controlled — LessWronglesswrong.com
- Simulators — LessWronglesswrong.com
- Building Technology to Drive AI Governancebounded-regret.ghost.io
- Recommendations for Technical AI Safety Research Directionsalignment.anthropic.com
- The Case Against AI Control Research — LessWronglesswrong.com
- Introducing Transluce | Transluce AItransluce.org
- A History of the Future, 2025-2040 — LessWronglesswrong.com
- AI Control: Improving Safety Despite Intentional Subversion — LessWronglesswrong.com
- Toward A Public Science of Model Behavior | Transluce AItransluce.org