~100 Interesting Questions

I saw the application for SPARC, a rationality summer program for high school students, earlier this year. It included a prompt I found super exciting: “List ~100 questions you have that interest you.” This is my response. (It’s not very organized—I go in no particular order and I sometimes interject some of my current thoughts about the questions.) I would love to hear people’s thoughts on any of these questions, and I would love to see other people post their responses.

  1. What does quantum entanglement mean for causality?

    1. Due to entanglement, there can be spacelike separated measurements such that there exists a reference frame where it looks like measurement A precedes and has a causal influence on the outcomes of measurement B, and also a reference frame where it looks like measurement B precedes and has a causal influence on the outcomes of measurement A.

    2. How do I rely on the notion of causality in my reasoning about the world? Do I have to revise this usage in practice?

  2. Why is mathematics so (unreasonably) effective in the natural sciences?

    1. Why is the universe structured in such a way as to have the same math accurately describe so many different aspects of the physical world?

  3. How can the Continuum Hypothesis be independent of the ZFC axioms? Why does the lack of “explicit” examples of sets with a cardinality between that of the naturals and that of the reals not guarantee that there are no examples at all? What would an “implicit” example even mean?

  4. What are the biggest easy wins in instrumental rationality?

    1. Candidates for me: Deleting the Youtube app, jumping into work immediately upon waking up (before brushing or eating or doing anything)

  5. Problem of Induction: Under what conditions do /​ should we expect the future to be like the past?

  6. What open problems only require theoretical genius to solve? (As opposed to slow and steady engineering progress.)

    1. Probably many problems in math and theoretical CS?

  7. Is my confidence in total hedonistic utilitarianism, despite the extent of its controversy among smart people, appropriate?

  8. What does it mean for something to be fundamental? To belong in a proper ontology of the universe? (Does it mean anything?)

  9. Is consciousness fundamental?

  10. What effects does consciousness have?

  11. How was consciousness incentivized by evolution, if at all? (I now think it probably was.)

  12. Is there a really good reason to believe or not believe one of the following theories of consciousness? (I think I find b most likely.) (I do not consider epiphenomenalism a serious option anymore, basically due to the arguments described here.) (I don’t really consider c a very serious option either.)

    1. Everything is physical—consciousness isn’t fundamental. (Basic response—I feel like any attempt to describe consciousness in terms of physical phenomena will be missing something.)

    2. Consciousness is fundamental. Conscious experiences are caused by physical processes, and have effects on physical processes in return. (Basic response—it seems a little absurd to believe that the Standard Model makes incorrect predictions in the brain because consciousness intervenes. It feels even more absurd to imagine there is any reasonable answer to the question, “How fast can physical processes affect consciousness, and how fast can consciousness affect physical processes?”)

    3. Hindu philosophy (poorly represented): We are vessels for a universal consciousness. (I kind of feel like there ought to be a knockdown argument against this, but I actually don’t think I’ve found one yet.) (Basic response—it sure seems like my consciousness is distinct from everyone else’s.)

  13. What beliefs about consciousness that I may hold (option b above?) involve weird latent laws of physics waiting for the physical structures that produce consciousness to arise in the universe? What is an appropriate response?

    1. Perhaps panpsychism is an appropriate response

  14. Do the laws of physics belong in a proper ontology of the universe?

  15. Is it appropriate to expect the laws of physics that have so far remained constant across time to continue to remain constant?

  16. What evidence can we get for physical systems, such as AI’s, being conscious?

  17. How should we approach the development of AI when it seems very hard to tell whether /​ to what extent AI is conscious at any given time?

  18. What can we know with absolute certainty?

    1. Maybe that my conscious experience does in fact consist of what it appears to consist of, mathematically proven statements, and tautologies? But maybe even those can be questioned?

  19. Does the universe exist? How seriously should I take various arguments against the universe’s existence?

    1. Solipsism, Boltzmann brains, Simulation argument

  20. What is the most reasonable response to the Fermi paradox?

  21. What kinds of anthropic reasoning are reasonable? What kinds are flawed?

  22. Did cosmic inflation really happen? Is the evidence sufficiently compelling to believe it? Has there been experimental confirmation of any predictions other than what the power spectrum of the CMB should look like, conditioned on the number of peaks? How wild is that prediction before using inflationary theory?

  23. Is the universe exactly flat? Is it appropriate to have a uniform prior on the cosmic density parameter (Omega)?

  24. How can a person intentionally become better at interacting with people?

  25. How does emotional intelligence develop? Can it be learned later in life?

  26. Why do neural networks generalize so well?

  27. What are the inductive biases of neural networks?

  28. Is deep learning enough for AGI?

  29. Will we ever truly enter a machine intelligence era? Or will the development of AI and its contributions to the world be more mundane than that?

  30. If we are going to enter a machine intelligence era, when will that happen?

  31. What do we mean when we use probabilities?

    1. Usually degree of belief, plus primitivism for probabilistic quantum phenomena?

  32. How should the insights of theoretical CS inform my general reasoning about the world?

  33. How can a person systematically improve the way they learn? What does optimal learning look like?

  34. What should I do?

    1. As a matter of general principle

    2. In my current circumstances

  35. What are the philosophical implications of Godel’s Incompleteness Theorems? Does truth transcend proof? In what sense? What other results from mathematical logic and computability theory are closely related, and what are their philosophical implications?

  36. How can we deliberately design the training procedures of neural networks to more reliably get them to behave how we want them to?

  37. How can we build Artificial General Intelligence /​ Artificial Superintelligence? (Not sure I actually want to know too soon, but it’s definitely interesting.)

  38. How did evolution lead to general intelligence in humans?

  39. How do humans learn representations?

  40. How much can human intuitions develop through new experiences?

    1. Could playing a well-designed video game (to the point of “knowing the meta”?) allow a human to have thorough intuitions about quantum mechanics? About 4D space?

  41. Do humans have the cognitive capacity to eventually do basically anything that is physically possible?

    1. How do we operationalize the notion of “physically possible” here? In some sense, we always do everything that is physically possible, since only one thing is actually possible, namely the laws of physics playing out. Maybe something like “locally physically possible,” where we don’t specify the entire state of the universe, just the state of a local region of it, and then verify that this local specification does not violate any local laws.

  42. What are we really doing when we talk about counterfactuals? Is there any actually principled way to consider them? If not, why does nothing go wrong in our standard use-cases for counterfactuals?

  43. What are some good things I could do to expand my comfort zone?

  44. What are the benefits of expanding one’s comfort zone?

  45. Why do people sometimes experience a certain mood with no clear context for why they should feel that way? (e.g. unexcited, sad) For negative ones, how can a person escape that feeling despite being unable to trace its source?

  46. How can a person kickstart themselves during a period of low motivation?

  47. What are unconventional social circumstances that would be highly beneficial to at least some people if they were to enter into them?

    1. Possible example: several couples with kids living together, sharing some responsibilities of raising the children.

  48. What are all the useful-to-isolate components of standard human functioning and intelligent human behavior?

    1. Particularly interesting when considered alongside AI systems

    2. Possible example: We have compact and reusable representations of high-level things, like tables, that we then use for causal reasoning.

  49. How can we build superintelligent AI systems that are extremely beneficial to the universe?

  50. Should we expect interactions with alien civilizations to be a critical part of the long-term future of human civilization? If so, how should we be thinking about it now?

  51. Is it reasonable to expect that we will be able to build interpretability tools that allow us to understand the largest and most powerful neural networks at a level that enables us to use this knowledge to steer them effectively?

  52. What are the most promising AI alignment strategies?

  53. Generally, what actions can be taken by what actors to most effectively increase the expected impact of AI?

  54. What AI policies can be implemented by what actors to most effectively increase the expected impact of AI?

  55. What people or organizations are going to build AGI?

  56. Do the empirical facts about the world today indicate that total hedonistic utilitarians should be longtermists?

  57. What are some stable equilibria in philosophical belief space? What concepts support these equilibria and enable lots of people to hold these beliefs?

  58. How can people (like me) who spend lots of time thinking about meta-level ideas keep from frequently sabotaging their own ability to “enjoy the moment?”

  59. To what extent is it possible to detach one’s well-being from their material circumstances?

  60. Is there an objective measure for how good a work of art is?

    1. Aggregate the opinions of all people

      1. Does aggregate mean take an average or a sum?

      2. All people who actually observe it?

      3. All people who exist?

      4. All possible minds?

      5. All people who “get it”?

  61. Problem of universals: Do universals exist independent of the particulars which instantiate them?

  62. Why can we talk about how long ago the Big Bang happened without running into major issues with the non-absolute-ness of time? Are we talking exclusively about our (Earth’s) reference frame? Are there other “reasonable” reference frames in which the age of the universe is very different?

  63. What philosophies can a person be made to buy into with a sufficiently well-written narrative?

    1. Authors tend to be able to make readers root for who they want them to root for (in books where there is someone we are supposed to root for).

  64. How good is the average human conscious experience right now?

  65. How good is the average conscious experience on Earth right now?

    1. Wild animal welfare is a major consideration.

  66. How good is the average human conscious experience over all time up to now?

  67. How good is the average conscious experience on Earth over all time up to now?

  68. How problematic is infinite ethics for a person (like me) who thinks the value of the universe is given by the sum over all conscious experiences ever of [the integral with respect to (subjective) time, from the first conscious moment to the last, of the conscious experience’s well-being]?[1]

  69. Is the universe infinite in space?

    1. Are there infinite conscious experiences currently ongoing?

    2. Are there infinite conscious experiences identical to my own currently ongoing?

  70. Is the universe infinite in time?

  71. What would I do if I were optimizing for my own happiness?

  72. What would I do if I were optimizing for intellectual stimulation?

  73. What would I do if I were optimizing for “success according to conventional definitions of success?”

  74. How should I adjust my choices in accord with the answers to the above three questions?

  75. What does it take to be able to look at the behavior of an agent and identify when that agent’s behavior fails to systematically achieve the agent’s own values?

    1. How would we create an AI that could learn human values without being confused by the fact that we often fail to systematically achieve our values?

  76. How important is it to maintain a diversity of cultures in the world?

  77. What is the maximum amount of productive work I can do in a sustainable manner?

    1. How can a person find this out?

    2. Should I really be aiming for this?

  78. What major insights about the world do other people have that I do not have? How can I acquire as many of these as possible?

  79. How can a person intentionally optimize their social environment to have as positive an influence on their life as possible?

  80. What is an appropriate way to assign prior probabilities to things?

  81. What constitutes a scientific theory? Are there distinct legitimate scientific theories that make exactly the same predictions? That is, is scientific theory underdetermined by evidence?

  82. How does one construct /​ deal with probability distributions over sets larger than the real numbers? (e.g. over the set of all possible ways the universe could be.)

  83. Does space exist independent of the material objects situated within it?

    1. Time probably doesn’t

  84. Are space and/​or time fundamental? Or are they emergent phenomena? What would that mean?

  85. Are there ongoing moral catastrophes?

  86. Does marginal reasoning plus the view that individuals cannot significantly shift the margins on large issues imply that individuals trying to maximize their impact should focus all their efforts towards one issue rather than diversifying? Is there any principled counterargument to this?

  87. Should I be paying more attention to things that fit into mainstream conceptions of where people can have a positive impact (e.g. politics, law, medicine, climate) that I find less interesting than utilitarian philosophy + AI X-risk?

  88. How can I efficiently learn more about what lives very different from my own (e.g. in many other countries) are like?

  89. What scientific and technological advancements could play a very large role in the long-term future of humanity? What does the path to achieving them look like?

    1. E.g. interstellar travel, ability to settle other planets, brain-computer interfaces

    2. The path to many of these may be, “Build artificial superintelligence, have ASI make it”

  90. Why is there something rather than nothing?

  91. How has evolution shaped human goals?

  92. What factors lead to variation in values across humans?

  93. Will AI’s undertake processes of value and goal formation that are comparable to those of humans?

  94. What are the most interesting questions on other people’s minds?

  95. What is a response commensurate to existing?

  96. How should the notion and inevitability of death inform the way I live?

  97. What are some characteristic qualities of exceptional works of art?

  98. Are there opportunities for having a massive positive impact on the world that no one has thought of yet?

  99. How good is the average conscious experience on factory farms today?

  100. What are trends like for quantity and quality of conscious experiences on factory farms?

  101. What is the most efficient way to convert money into the well-being of conscious creatures?


  1. Roughly answered:

    1. Do we have free will?

  2. I don’t know or remember enough about these topics to really know what the specific interesting questions are:

    1. What’s the deal with UDASSA? Should I actually not think of heat death as a critical component in my estimation of the value of the universe?

    2. What’s the deal with quantum gravity?

    3. What’s the deal with the black hole information paradox?

  1. ^

    The value of the universe: