full transcript
From the Ted Talk by Janelle Shane: The danger of AI is weirder than you think
Unscramble the Blue Letters
So let's say that you were trying to get an AI to take this cootecllin of robot parts and assemble them into some kind of robot to get from Point A to Point B. Now, if you were going to try and solve this problem by writing a traditional-style computer program, you would give the program step-by-step instructions on how to take these parts, how to assemble them into a robot with legs and then how to use those legs to walk to Point B. But when you're using AI to solve the problem, it goes differently. You don't tell it how to solve the prelobm, you just give it the goal, and it has to figure out for itself via trial and error how to raech that goal. And it turns out that the way AI tends to slove this particular problem is by doing this: it alssmeebs itself into a tower and then flals over and lands at piont B. And technically, this solves the problem. Technically, it got to Point B. The danger of AI is not that it's going to rebel against us, it's that it's going to do exactly what we ask it to do. So then the tcrik of working with AI becomes: How do we set up the problem so that it actually does what we want?
Open Cloze
So let's say that you were trying to get an AI to take this __________ of robot parts and assemble them into some kind of robot to get from Point A to Point B. Now, if you were going to try and solve this problem by writing a traditional-style computer program, you would give the program step-by-step instructions on how to take these parts, how to assemble them into a robot with legs and then how to use those legs to walk to Point B. But when you're using AI to solve the problem, it goes differently. You don't tell it how to solve the _______, you just give it the goal, and it has to figure out for itself via trial and error how to _____ that goal. And it turns out that the way AI tends to _____ this particular problem is by doing this: it _________ itself into a tower and then _____ over and lands at _____ B. And technically, this solves the problem. Technically, it got to Point B. The danger of AI is not that it's going to rebel against us, it's that it's going to do exactly what we ask it to do. So then the _____ of working with AI becomes: How do we set up the problem so that it actually does what we want?
Solution
- assembles
- collection
- solve
- point
- falls
- reach
- trick
- problem
Original Text
So let's say that you were trying to get an AI to take this collection of robot parts and assemble them into some kind of robot to get from Point A to Point B. Now, if you were going to try and solve this problem by writing a traditional-style computer program, you would give the program step-by-step instructions on how to take these parts, how to assemble them into a robot with legs and then how to use those legs to walk to Point B. But when you're using AI to solve the problem, it goes differently. You don't tell it how to solve the problem, you just give it the goal, and it has to figure out for itself via trial and error how to reach that goal. And it turns out that the way AI tends to solve this particular problem is by doing this: it assembles itself into a tower and then falls over and lands at Point B. And technically, this solves the problem. Technically, it got to Point B. The danger of AI is not that it's going to rebel against us, it's that it's going to do exactly what we ask it to do. So then the trick of working with AI becomes: How do we set up the problem so that it actually does what we want?
Frequently Occurring Word Combinations
ngrams of length 2
collocation |
frequency |
ice cream |
2 |
human fingers |
2 |
recognize trucks |
2 |
Important Words
- ai
- assemble
- assembles
- collection
- computer
- danger
- differently
- error
- falls
- figure
- give
- goal
- instructions
- kind
- lands
- legs
- parts
- point
- problem
- program
- reach
- rebel
- robot
- set
- solve
- solves
- technically
- tower
- trial
- trick
- turns
- walk
- working
- writing