Machine Learning Street Talk (MLST)
ROBERT MILES - "There is a good chance this kills everyone"
Please check out Numerai - our sponsor @
https://num...
more
May 21 2023 2h 1m
Chapter 1 2 mins
IntroChapter 2 2 mins
Numerai Sponsor MesssageChapter 3 13 mins
AI AlignmentChapter 4 5 mins
Limits of AI Capabilities and PhysicsChapter 5 7 mins
AI Progress and TimelinesChapter 6 7 mins
AI Arms Race and InnovationChapter 7 4 mins
Human-Machine Hybrid IntelligenceChapter 8 7 mins
Understanding and Defining IntelligenceChapter 9 13 mins
AI in Conflict and Cooperation with HumansChapter 10 2 mins
Interpretability and Mind Reading in AIChapter 11 1 min
Mechanistic Interpretability and Deconfusion ResearchChapter 12 2 mins
Understanding the core concepts of AIChapter 13 1 min
Moon landing analogy and AI alignmentChapter 14 4 mins
Cognitive horizon and limits of human intelligenceChapter 15 2 mins
Funding and focus on AI alignmentChapter 16 7 mins
Regulating AI technology and potential risksChapter 17 2 mins
Aligning AI with human values and its dynamic natureChapter 18 3 mins
Cooperation and AllyshipChapter 19 2 mins
Orthogonality Thesis and Goal PreservationChapter 20 55 sec
Anthropomorphic Language and Intelligent AgentsChapter 21 2 mins
Maintaining Variety and Open-ended ExistenceChapter 22 4 mins
Emergent Abilities of Large Language ModelsChapter 23 5 mins
Convergence vs EmergenceChapter 24 3 mins
Criticism of X-risk and Alignment CommunitiesChapter 25 37 sec
Fusion of AI communities and addressing biasesChapter 26 53 sec
AI systems integration into society and understanding themChapter 27 23 sec
Changing opinions on AI topics and learning from past videosChapter 28 3 mins
Utility functions and von Neumann-Morgenstern theoremsChapter 29 2 mins
AI Safety FAQ projectChapter 30 1 min
Building a conversation agent using AI safety dataset