All pages
- 3Blue1Brown
- AI polarization
- AI prepping
- AI safety field consensus
- AI safety is harder than most things
- AI safety is not a community
- AI safety lacks a space to ask stupid or ballsy questions
- AI safety technical pipeline does not teach how to start having novel thoughts
- AI takeoff
- AI timelines
- AI will solve everything argument against AI safety
- Add all permutations of a card to prevent pattern-matching
- Add the complete proof on proof cards to reduce friction when reviewing
- Agent foundations
- AlphaGo
- AlphaGo Zero
- AlphaGo as evidence of discontinuous takeoff
- Analyzing disagreements
- Andy Matuschak
- Anki deck options
- Anki deck philosophy
- Anki reviews are more fun on mobile
- Architecture
- Asymmetric institution
- Asymmetric weapon
- Asynchronous support
- Big card
- Booster card
- Braid for math
- Broad augmentation
- Bury cards to speed up review
- Busy life periods and spaced inbox
- Can spaced repetition interfere with internal sense of relevance?
- Card sharing
- Card sharing allows less valuable cards to be created
- Cards created by oneself can be scheduled more aggressively
- Carl
- Carl Shulman
- Central node trick for remembering equivalent properties
- Changing selection pressures argument
- Choosing problems for spaced proof review
- Christiano's operationalization of slow takeoff
- Coherence and goal-directed agency discussion
- Combinatorial explosion in math
- Comparison of AI takeoff scenarios
- Comparison of sexually transmitted diseases
- Comparison of terms related to agency
- Competence gap
- Content sharing between AIs
- Continuous AI takeoff
- Continuous takeoff
- Continuous takeoff keyhole
- Convergent evolution of values
- Corrigibility
- Corrigibility may be undesirable
- Counterfactual of dropping a seed AI into a world without other capable AI
- Creative forgetting
- DSA
- Dealing with bad problems in spaced proof review
- Debates shift bystanders' beliefs
- Deck options for proof cards
- Deck options for small cards
- Deconfusion
- Depictions of learning in The Blue Lagoon are awful
- Desiderata for cognitive reductions
- Different mental representations of mathematical objects is a blocker for an exploratory medium of math
- Different senses of claims about AGI
- Discontinuities in usefulness of whole brain emulation technology
- Discontinuities in whole brain emulation technology
- Discursive texts are difficult to ankify
- Distillation is not enough
- Distortion of true frequency of piece of knowledge due to spaced repetition
- Do an empty review of proof cards immediately after adding to prevent backlog
- Dual ratings for spaced inbox
- Duolingo
- Duolingo does repetition at the lesson level
- Duolingo for math
- Education in The Blue Lagoon is awful
- Eliezer
- Emotional difficulties of AI safety research
- Emotional difficulties of spaced repetition
- Empty review
- Encoding dependence problem
- Equivalence classes of prompts
- Evolution
- Evolution analogy
- Existential win
- Existing implementations of card sharing have nontrivial overhead
- Expert response heuristic for prompt writing
- Explorable explanation
- Fake review
- Fast takeoff
- Feeling like a perpetual student in a subject due to spaced repetition
- Feynman technique fails when existing explanations are bad
- Finiteness assumption in explorable media
- Flag things to fix during review
- Fractional progress argument for AI timelines
- Future planning
- Generic advice is difficult to give but also important
- Giving advice in response to generic questions is difficult but important
- Goal inference
- Goalpost for usefulness of HRAD work
- Guided by the beauty of our weapons
- HCH
- HRAD
- Hansonian
- Hardware-driven vs software-driven progress
- Hardware argument for AI timelines
- Hardware overhang
- Highly reliable agent designs
- Hnous
- Hnous927
- How doomed are ML safety approaches?
- How meta should AI safety be?
- Human safety
- Human safety problem
- Human safety problems
- Humans consulting HCH
- Hyperbolic growth
- IDA
- Ignore Anki add-ons to focus on fundamentals
- Importance of knowing about AI takeoff
- Improvement curve for good people
- Incremental reading
- Incremental reading in Anki
- Integration card
- Intelligence amplification
- Inter-personal comparison test
- Interacting with copies of myself
- Interaction reversal between knowledge-to-be-memorized and ideas-to-be-developed
- Interaction reversal between knowledge and notes
- Intra-personal comparison test
- Is AI safety no longer a scenius?
- It is difficult to find people to bounce ideas off of
- It is difficult to get feedback on published work
- Iterated amplification
- Iteration cadence for spaced repetition experiments
- Jelly no Puzzle
- Jessica Taylor
- Jonathan Blow
- Kanzi
- Kasparov Window
- Kasparov window
- Keyhole argument against continuous takeoff
- Laplace's rule of succession argument for AI timelines
- Large graduating interval as a way to prevent pattern-matching
- Large graduating interval as substitute for putting effort into making atomic cards
- Late singularity
- Learning-complete
- Learning complete
- Linked list proof card
- List of AI safety projects I could work on
- List of arguments against working on AI safety
- List of big discussions in AI alignment
- List of breakthroughs plausibly needed for AGI
- List of critiques of iterated amplification
- List of disagreements in AI safety
- List of experiments with Anki
- List of men by number of sons, daughters, and wives
- List of people who have thought a lot about spaced repetition
- List of reasons something isn't popular or successful
- List of success criteria for HRAD work
- List of teams at OpenAI
- List of technical AI alignment agendas
- List of techniques for making small cards
- List of techniques for managing working memory in explanations
- List of terms used to describe the intelligence of an agent
- List of thought experiments in AI safety
- List of timelines
- List of timelines for futuristic technologies
- Live math video
- Lumpiness
- Lumpy
- MIRI
- MIRI vs Paul research agenda hypotheses
- Main Page
- Maintaining habits is hard, and spaced repetition is a habit
- Make Anki cards based on feedback you receive
- Make new cards when you get stuck
- Managing micro-movements in learning
- Mapping mental motions to parts of a spaced repetition algorithm
- Mass shift to technical AI safety research is suspicious
- Master How To Learn guy
- Math vs video games
- Medium that reveals flaws
- Medium that reveals its flaws
- Meta-execution
- Michael Nielsen
- Minimal AGI vs task AGI
- Missing gear
- Missing gear for intelligence
- Missing gear vs secret sauce
- Mixed messaging regarding independent thinking
- My current thoughts on the technical AI safety pipeline (outside academia)
- My take on RAISE
- My understanding of how IDA works
- Narrow augmentation
- Narrow vs broad cognitive augmentation
- Narrow window argument against continuous takeoff
- Nate
- Newcomers in AI safety are silent about their struggles
- Nobody understands what makes people snap into AI safety
- One-sentence summary card
- One wrong number problem
- Ongoing friendship and collaboration is important
- Online question-answering services are unreliable
- Open-ended questions are common in real life
- OpenAI
- Optimal unlocking mechanism for booster cards is unclear
- Output curve for good people
- Paperclip maximizer
- Parasitizing on popularity
- Pascal's mugging and AI safety
- Pascal's mugging argument against AI safety
- Paul
- Paul Christiano
- Paul christiano
- People are bad
- People watching
- Personhood API vs therapy axis of interpersonal interactions
- Physical vs digital clutter
- Piotr Wozniak
- Politicization of AI
- Popularity symbiosis
- Probability and statistics as fields with an exploratory medium
- Progress in self-improvement
- Proof card
- Prosaic AI
- Quick review
- Quotability vs ankifiability
- RCG
- RSI
- Rapid capability gain vs AGI progress
- Realism about rationality
- Reference class forecasting on human achievements argument for AI timelines
- Repetition granularity
- Representing impossibilities
- Resource overhang
- Reverse side card for everything
- Richard Ngo
- Scaling hypothesis
- Scenius
- Science argument
- Second species argument
- Secret sauce for intelligence
- Secret sauce for intelligence vs specialization in intelligence
- Selection effect for successful formalizations
- Selection effect for who builds AGI
- Self-graded prompts made for others must provide guidance for grading
- Setting up Windows
- Short-term preferences-on-reflection
- Should booster cards be marked as new?
- Simple core
- Simple core algorithm
- Simple core algorithm for agency
- Simple core algorithm of agency
- Simple core of consequentialist reasoning
- Slow feedback cycle for spaced repetition
- Slow feedback loop for spaced repetition
- Slow takeoff
- Small card
- Snoozing epicycle
- Soft-hard takeoff
- Soft alarm clock
- Soft takeoff keyhole
- Something like realism about rationality
- Soren Bjornstad
- Spaced everything
- Spaced inbox
- Spaced inbox ideas
- Spaced proof review
- Spaced proof review as a way to invent novel proofs
- Spaced proof review as a way to understand key insights in a proof
- Spaced proof review is not about memorizing proofs
- Spaced proof review routine
- Spaced repetition
- Spaced repetition allows graceful deprecation of experiments
- Spaced repetition and cleaning one's room
- Spaced repetition as generator of questions
- Spaced repetition as soft alarm clock
- Spaced repetition constantly reminds one of inadequacies
- Spaced repetition experiments take months to complete
- Spaced repetition is not about memorization
- Spaced repetition is useful because most knowledge is sparsely applicable
- Spaced repetition isn't about memorization
- Spaced repetition prevents unrecalled unrecallables
- Spaced repetition world
- Spaced writing inbox
- Spoiler test of depth
- Statistical analysis of expert timelines argument for AI timelines
- Steam game buying algorithm
- Stream of low effort questions helps with popularity
- Stupid questions
- Sudden emergence
- Summary of my beliefs
- Switching costs of various kinds of software
- Takeoff speed
- Takeoff speeds
- Tao Analysis I exercise count
- Tao Analysis Solutions
- Task-dependent diversity
- Teaching-complete
- Teaching complete
- Terms used to describe intelligence of agents
- Test
- Text to speech software
- The Hour I First Believed
- The Precipice notes
- The Secret of Psalm 46
- The Secret of Psalm 46 outline
- The Sequences vs evergreen notes
- The Uncertain Future
- The Witness
- The mathematics community has no clear standards for what a mathematician should know
- There is pressure to rush into a technical agenda
- There is room for something like RAISE
- Thinking Mathematics
- Thinking Physics for math
- Timeline of my involvement in AI safety
- Tinkering in math requires loading the situation into working memory
- Toby Ord questions
- Tutoring heuristic for prompt writing
- UDASSA
- Unbounded working memory assumption in explanations
- Uncertain Future
- Understanding is discontinuous
- Uninsightful articles can seem insightful due to unintentional spaced repetition
- Unintended consequences of AI safety advocacy argument against AI safety
- Unreliability of online question-answering services makes it emotionally taxing to write up questions
- Use paper during spaced repetition reviews
- Use temporary separate Anki decks to learn new cards based on priority
- Using Anki for math
- Using spaced repetition to improve public discourse
- Value learning
- Video games allow immediate exploration
- Video games comparison to math
- Video games vs math
- Video games work with finite and discrete objects
- Vow of silence
- WBE
- Website to aggregate solutions to textbook exercises
- Wei
- Wei Dai
- What would a vow of silence look like for math?
- Whole brain emulation
