The blue-minimising robot and model splintering — AI Alignment Forum
alignmentforum.org · 1,724 words · saved by 1 readers
A long time ago, Scott introduced the blue-minimising robot: …
x The blue-minimising robot and model splintering — AI Alignment Forum AI Frontpage 6 The blue-minimising robot and model splintering by Stuart_Armstrong 28th May 2021 4 min read 5 6 A long time ago, Scott introduced the blue-minimising robot : Imagine a robot with a turret-mounted camera and laser. Each moment, it is programmed to move forward a certain distance and perform a sweep with its camera. As it sweeps, the robot continuously analyzes the average RGB value of the pixels in the camera image; if the blue component passes a certain threshold, the robot stops, fires its laser at the part
related reading
- A Pragmatic Vision for Interpretability — AI Alignment Forumalignmentforum.org
- Optimality is the tiger, and agents are its teeth — LessWronglesswrong.com
- The Stamp Collectormindingourway.com
- Deep Deceptiveness — LessWronglesswrong.com
- Embedded Agency (full-text version) — LessWronglesswrong.com
- A positive case for how we might succeed at prosaic AI alignment — AI Alignment Forumalignmentforum.org
- What failure looks like — AI Alignment Forumalignmentforum.org
- A Pragmatic Vision for Interpretability — LessWronglesswrong.com
- How Can Interpretability Researchers Help AGI Go Well? — AI Alignment Forumalignmentforum.org
- Beliefs are Chosen to Serve Goals — LessWronglesswrong.com
- Mediumdeepmindsafetyresearch.medium.com
- Reward Is Not Enough — LessWronglesswrong.com