Expand each heading to view my notes.

The aim is to understand

• What are foundational models? How are they trained?
• What is the current state of the art in ML?
• What are the capability requirements for an AI to be an AGI?
• Contrast modern ML with your blueprint for AGI and evaluate if and when it could be developed.

Progress of Language Models

• https://theaidigest.org/progress-and-dangers - progress of AI language models

Foundational Models

• https://arxiv.org/abs/2108.07258 - On the opportunities and risks of foundational models

Four premises/claims behind MIRI’s mission

• https://intelligence.org/2015/07/24/four-background-claims/

AGI Safety from first principles

• https://www.alignmentforum.org/s/mzgtmmTKKn5MuCzFJ/p/8xRSjC76HasLnMGSf

Why and how of scaling LLMs

• https://www.alignmentforum.org/posts/midXmMb2Xg37F2Kgn/new-scaling-laws-for-large-language-models

Future ML Systems will be qualitatively different

• https://bounded-regret.ghost.io/future-ml-systems-will-be-qualitatively-different/

More is Different for AI

• https://bounded-regret.ghost.io/more-is-different-for-ai/