full transcript

From the Ted Talk by Nick Bostrom: What happens when our computers get smarter than we are?


Unscramble the Blue Letters


This can happen, and the outcome could be very good for humanity. But it doesn't happen automatically. The initial conditions for the intelligence explosion might need to be set up in just the right way if we are to have a controlled detonation. The values that the A.I. has need to macth ours, not just in the fiaiamlr context, like where we can easily check how the A.I. behaves, but also in all novel contexts that the A.I. might eeonuntcr in the indefinite future.

And there are also some esoteric issues that would need to be solved, steord out: the exact details of its decision theory, how to deal with laogcil uncertainty and so forth. So the technical problems that need to be solved to make this work look quite difficult — not as difficult as mnikag a superintelligent A.I., but fairly difficult. Here is the worry: Making sreupngeniltleit A.I. is a really hard challenge. Making superintelligent A.I. that is safe iloenvvs some additional challenge on top of that. The risk is that if somebody figures out how to crack the first challenge without also having cracked the additional challenge of ensuring perfect safety.

Open Cloze


This can happen, and the outcome could be very good for humanity. But it doesn't happen automatically. The initial conditions for the intelligence explosion might need to be set up in just the right way if we are to have a controlled detonation. The values that the A.I. has need to _____ ours, not just in the ________ context, like where we can easily check how the A.I. behaves, but also in all novel contexts that the A.I. might _________ in the indefinite future.

And there are also some esoteric issues that would need to be solved, ______ out: the exact details of its decision theory, how to deal with _______ uncertainty and so forth. So the technical problems that need to be solved to make this work look quite difficult — not as difficult as ______ a superintelligent A.I., but fairly difficult. Here is the worry: Making ________________ A.I. is a really hard challenge. Making superintelligent A.I. that is safe ________ some additional challenge on top of that. The risk is that if somebody figures out how to crack the first challenge without also having cracked the additional challenge of ensuring perfect safety.

Solution


  1. sorted
  2. familiar
  3. logical
  4. involves
  5. encounter
  6. superintelligent
  7. match
  8. making

Original Text


This can happen, and the outcome could be very good for humanity. But it doesn't happen automatically. The initial conditions for the intelligence explosion might need to be set up in just the right way if we are to have a controlled detonation. The values that the A.I. has need to match ours, not just in the familiar context, like where we can easily check how the A.I. behaves, but also in all novel contexts that the A.I. might encounter in the indefinite future.

And there are also some esoteric issues that would need to be solved, sorted out: the exact details of its decision theory, how to deal with logical uncertainty and so forth. So the technical problems that need to be solved to make this work look quite difficult — not as difficult as making a superintelligent A.I., but fairly difficult. Here is the worry: Making superintelligent A.I. is a really hard challenge. Making superintelligent A.I. that is safe involves some additional challenge on top of that. The risk is that if somebody figures out how to crack the first challenge without also having cracked the additional challenge of ensuring perfect safety.

Frequently Occurring Word Combinations


ngrams of length 2

collocation frequency
artificial intelligence 5
optimization process 3
control problem 3
ed witten 2
intelligence explosion 2
machine intelligence 2
mathematical problem 2
powerful optimization 2
human hackers 2
making superintelligent 2
additional challenge 2

ngrams of length 3

collocation frequency
powerful optimization process 2


Important Words


  1. additional
  2. automatically
  3. behaves
  4. challenge
  5. check
  6. conditions
  7. context
  8. contexts
  9. controlled
  10. crack
  11. cracked
  12. deal
  13. decision
  14. details
  15. detonation
  16. difficult
  17. easily
  18. encounter
  19. ensuring
  20. esoteric
  21. exact
  22. explosion
  23. familiar
  24. figures
  25. future
  26. good
  27. happen
  28. hard
  29. humanity
  30. indefinite
  31. initial
  32. intelligence
  33. involves
  34. issues
  35. logical
  36. making
  37. match
  38. outcome
  39. perfect
  40. problems
  41. risk
  42. safe
  43. safety
  44. set
  45. solved
  46. sorted
  47. superintelligent
  48. technical
  49. theory
  50. top
  51. uncertainty
  52. values
  53. work