OpenAI's new ChatGPT explains the worst-case time complexity of the bubble sort algorithm, with Python code examples, in the style of a fast-talkin' wise guy from a 1940's gangster movie:

Dec 1, 2022 · 1:39 AM UTC

201
2,934
546
19,029
nitter.net/pomothug/status/…
1
691
Replying to @goodside
This thing is pretty impressive. But can also make silly mistakes
1
20
In general, you shouldn't expect it to be able to perform accurate calculations, at least not "in its head" — this is a known limitation of models like GPT-3. It only stands a chance if prompted to write calculations out step-by-step like one might do on paper.
1
2
69
more replies
Replying to @goodside
I like how the comments in the python code are *not* in 1940s gangster style. Guess that would be a little too far out of distribution
1
17
You can sort of watch the code stylistically contaminating its speech — it starts with a comment that contains “ain’t”, but the convention for comments to have formal (or at least modern) style wins out once the actual code appears.
61
Replying to @goodside
How come this happens?
1
4
Tokenization. It doesn’t see text as sequences of characters, but of variable-length tokens about 4 chars on average. It generally can’t do anything that needs to be done one letter at a time. It’s also bad at counting in its head.
1
11
more replies
Replying to @goodside
But this *isn’t* the worst case! the unsorted list that it starts with should’ve been [8, 5, 4, 2, 1] I wouldn’t expect it to be able to do that… because it’s missing understanding Still astonishing! Take-home exams may be disappearing— Education forever changed!
4
102
Oh, so we got a wise guy, huh?
3
318
more replies
Replying to @goodside @punk6529
That's incredible!
1
Load more