How can we solve diffuse threats like research sabotage with AI control? — LessWrong

Misaligned AIs might engage in research sabotage: making safety research go poorly by doing things like withholding their best ideas or putting subtl…

x How can we solve diffuse threats like research sabotage with AI control? — LessWrong AI Control AI Frontpage 54 How can we solve diffuse threats like research sabotage with AI control? by Vivek Hebbar 30th Apr 2025 AI Alignment Forum 10 min read 1 54 Ω 34 Misaligned AIs might engage in research sabotage: making safety research go poorly by doing things like withholding their best ideas or putting subtle bugs in experiments. To mitigate this risk with AI control, we need very different techniques than those in the original control paper or Ctrl-Z , because research sabotage differs from those

saved by

• Emil Ryd

related reading

• How can we solve diffuse threats like research sabotage with AI control?blog.redwoodresearch.org
• How can we solve diffuse threats like research sabotage with AI control?redwoodresearch.substack.com
• The case for ensuring that powerful AIs are controlled — LessWronglesswrong.com
• Deep Deceptiveness — LessWronglesswrong.com
• The Case Against AI Control Research — LessWronglesswrong.com
• AI Control: Improving Safety Despite Intentional Subversion — LessWronglesswrong.com
• AI Control: Improving Safety Despite Intentional Subversion — AI Alignment Forumalignmentforum.org
• Recommendations for Technical AI Safety Research Directionsalignment.anthropic.com
• Thoughts on the conservative assumptions in AI controlblog.redwoodresearch.org
• Agentic misalignment: How LLMs could be insider threats \ Anthropicanthropic.com
• A basic systems architecture for AI agents that do autonomous research — LessWronglesswrong.com
• Notes on handling non-concentrated failures with AI control: high level methods and different regimesblog.redwoodresearch.org