QNR prospects are important for AI alignment research — LessWrong

Attention conservation notice: This discussion is intended for readers with an interest in prospects for knowledge-rich intelligent systems and poten…

x QNR prospects are important for AI alignment research — LessWrong Interpretability (ML & AI) Language Models (LLMs) AI Frontpage 94 QNR prospects are important for AI alignment research by Eric Drexler 3rd Feb 2022 AI Alignment Forum 13 min read 12 94 Ω 40 Attention conservation notice: This discussion is intended for readers with an interest in prospects for knowledge-rich intelligent systems and potential applications of improved knowledge representations to AI capabilities and alignment. It contains no theorems. Abstract Future AI systems will likely use quasilinguistic neural representat

related reading

• QNRs_FHI-TR-2021-3.0.pdfowainevans.github.io
• Alignment remains a hard, unsolved problem — LessWronglesswrong.com
• The Iliad Intensive Course Materials — LessWronglesswrong.com
• GenAI Handbookgenai-handbook.github.io
• A Pragmatic Vision for Interpretability — AI Alignment Forumalignmentforum.org
• Interpretability Dreamstransformer-circuits.pub
• [2005.11401] Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasksarxiv.org
• Automated Alignment Researchers: Using large language models to scale scalable oversight \ Anthropicanthropic.com
• Is AI Reasoning Right for the Wrong Reasons? | Quanta Magazinequantamagazine.org
• How AI Is Learning to Think in Secretnickandresen.substack.com
• How Can Interpretability Researchers Help AGI Go Well? — AI Alignment Forumalignmentforum.org
• Research | AI Alignment Foundationaialignmentfoundation.org