On AI Safety Jargon - by Benjamin Arnold - Turning Flukes
turningflukes.substack.com · 1,780 words · saved by 1 readers
Or, why there’s no need to say ‘heuristics’.
Paul Klee, 1939 (also me at times) When I was first getting acquainted with AI safety, one of the early challenges I encountered was the lexicon. I’m not from the EA-Rationalist-LessWrong-osphere; my prior familiarity with that world consisted mostly of knowing that Sam Bankman-Fried got Tom Brady to help him do fraud. The jargon was a shift; coming from a corporate background, I’m used to building the plane as we fly it, people moving my cheese, and servant leaders iterating on features that delight. When I started doing my AI reading, I found myself - like anyone else entering a new field…
saved by
related reading
- AI Safety for Fleshy Humans: a whirlwind touraisafety.dance
- The Artificiality of Alignmentjoinreboot.org
- The Case Against AI Control Research — LessWronglesswrong.com
- AI Control: Improving Safety Despite Intentional Subversion — LessWronglesswrong.com
- AI Safety Seems Hard to Measurecold-takes.com
- AI Safety: A Short FAQ for Mathematiciansalkjash.github.io
- Brain dump on the diversity of AI risk | Jascha’s blogsohl-dickstein.github.io
- Ten AI safety projects I'd like people to work onthirdthing.ai
- What is AI alignment? - by Adam Jones - BlueDot Impactblog.bluedot.org
- EA Safetycontraptions.venkateshrao.com
- Using Dangerous AI, But Safely?youtube.com
- Many arguments for AI x-risk are wrong — AI Alignment Forumalignmentforum.org