I bet that for the second random number in the same session, it is significantly less likely for an LLM to repeat its first number compared to two random draws. LLMs seem to mimic the human tendency to consider 7 as the most random, and I feel like repeating a random number would be perceived as not random.

The random numbers seem to be really stable on the first prompts!

For example:

pick a number between 1 - 10000

> I’ll go with 7,284.

Yeah I got 7284 as well on the first try. My second session got 7384.

ah, got 7421 too. I then it retry and got 7429.

me > pick a number between 1 to 10000

chatgpt > 7429

me > another one

chatgpt > 1863