Category:AI safety
From Issawiki (static archive)
For pages related to AI safety and alignment. These pages should be moved to a dedicated AI safety wiki at some point.
Pages in this category (116)
- Agent foundations
- AI prepping
- AI safety field consensus
- AI takeoff
- AI timelines
- AI will solve everything argument against AI safety
- Aligning smart AI using slightly less smart AI
- AlphaGo
- AlphaGo as evidence of discontinuous takeoff
- Analyzing disagreements
- Architecture
- Asymmetric institution
- Can the behavior of approval-direction be undefined or random?
- Carl Shulman
- Changing selection pressures argument
- Christiano's operationalization of slow takeoff
- Coherence and goal-directed agency discussion
- Comparison of AI takeoff scenarios
- Comparison of terms related to agency
- Competence gap
- Content sharing between AIs
- Continuous takeoff
- Corrigibility
- Corrigibility may be undesirable
- Counterfactual of dropping a seed AI into a world without other capable AI
- Debates shift bystanders' beliefs
- Deconfusion
- Different senses of claims about AGI
- Difficulty of AI alignment
- Discontinuities in usefulness of whole brain emulation technology
- Doomer argument against AI safety
- Evolution
- Existential win
- Explosive aftermath
- Fractional progress argument for AI timelines
- Goalpost for usefulness of HRAD work
- Hardware argument for AI timelines
- Hardware overhang
- Hardware-driven vs software-driven progress
- HCH
- Highly reliable agent designs
- How doomed are ML safety approaches?
- How similar are human brains to chimpanzee brains?
- Human safety problem
- Hyperbolic growth
- Importance of knowing about AI takeoff
- Intelligence amplification
- Iterated amplification
- Jessica Taylor
- Kanzi
- Kasparov window
- Laplace's rule of succession argument for AI timelines
- Late 2021 MIRI conversations
- Late singularity
- 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 success criteria for HRAD work
- List of teams at OpenAI
- List of technical AI alignment agendas
- List of terms used to describe the intelligence of an agent
- List of thought experiments in AI safety
- Lumpiness
- Meta-execution
- Minimal AGI
- Minimal AGI vs task AGI
- MIRI vs Paul research agenda hypotheses
- Missing gear for intelligence
- Missing gear vs secret sauce
- My understanding of how IDA works
- Narrow window argument against continuous takeoff
- Number of relevant actors around the time of creation of AGI
- One wrong number problem
- OpenAI
- Paperclip maximizer
- Pascal's mugging and AI safety
- Paul Christiano
- Philosophical difficulty
- Pivotal act
- Politicization of AI
- Progress in self-improvement
- Prosaic AI
- Rapid capability gain vs AGI progress
- Reference class forecasting on human achievements argument for AI timelines
- Resource overhang
- Richard Ngo
- Robin Hanson
- Scaling hypothesis
- 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
- Short-term preferences-on-reflection
- Simple core
- Simple core of consequentialist reasoning
- Single-architecture generality
- Single-model generality
- Soft-hard takeoff
- Something like realism about rationality
- Statistical analysis of expert timelines argument for AI timelines
- Stupid questions
- Sudden emergence
- Textbook test for AI theory
- The Uncertain Future
- UDASSA
- Unintended consequences of AI safety advocacy argument against AI safety
- Value learning
- Wei Dai
- Weird recursion
- Whole brain emulation
- Will it be possible for humans to detect an existential win?
- Will there be significant changes to the world prior to some critical AI capability threshold being reached?