Ok but then the question is: why would AI pick two numbers, out of 1 to 25? Should that not be evenly distributed instead? Why is the AI so skewed?
I believe though it is an old phenomenom that if (given the default context) you ask a random number, it will more often than not give numbers from a very small set.
I just used a "Temporary chat" with ChatGPT twice to "Give a random number" and it gave 42 both times (this shuold be rather understandable why it ended up with this). Then I asked "Give a random number in 1-10" twice and it responded with 7 for both (and this too is an old "magic number").
btw today's models run random.randint() by default unless you prohibit python by choosing one of the dumb models, disable thinking, disable tool calls, etc.
We are talking about 2 trials, right? Or did they do more repetitions?
What would be the "correct" numbers to get from 2 trials that show an even distribution?
---
For comparison:
I clicked twice to generate numbers on random.org
The second generated numbers were
Here are your random numbers:
13 22
Timestamp: 2026-10-27 10:01:22 UTC
> why would ~~AI~~ random.org pick two numbers, out of 1 to 25? Should that not be evenly distributed instead? Why is the AI so skewed?
(The first set of numbers were 20 and 12. Unfortunately I didn't get lucky to get 13/22 on the first try.)