An Argument Against AI Alignment — Kate Tolo
Everything you have ever built has been slowly falling apart. Some structures persist for millennia. The difference is anti-entropic systems — and they reveal why AI alignment will fail.
Everything you have ever built... every company, every relationship, every project, every version of yourself... has been slowly falling apart the entire time you've been building it. That familiar feeling of constant maintenance. The sense that if you look away for too long, something unravels. Yet, some structures don't seem to have this problem. Christianity has been alive for 2,000 years. Nobody is maintaining it. Nobody is holding it together through sheer will. It simply... persists. The Catholic Church has outlasted every empire, every corporation, every government that existed when it
saved by
related reading
- Kate Tolo (@_katetolo) on Xx.com
- What failure looks like — LessWronglesswrong.com
- AGI Ruin: A List of Lethalities — LessWronglesswrong.com
- What failure looks like — AI Alignment Forumalignmentforum.org
- Gradual Disempowermentgradual-disempowerment.ai
- ⿻ Symbiogenesis vs. Convergent Consequentialism — LessWronglesswrong.com
- Another (outer) alignment failure story — AI Alignment Forumalignmentforum.org
- The Problem — LessWronglesswrong.com
- Core views on AI safety: When, why, what, and how \ Anthropicanthropic.com
- The Artificiality of Alignmentjoinreboot.org
- Survival without dignity - Rudolf's Writingrudolf.website
- What is AI alignment? - by Adam Jones - BlueDot Impactblog.bluedot.org