[hx2a]

And yet, generative AI also seems to be poor at randomness. When I ask Google Gemini for a list of 50 random words, it gave me a list of 18 unique words, with 16 of them repeated exactly 3 times.

Abyss: 1 Ambiguous: 3 Cacophony: 3 Crescendo: 3 Ephemeral: 3 Ethereal: 3 Euphoria: 3 Labyrinth: 3 Maverick: 3 Melancholy: 3 Mellifluous: 3 Nostalgia: 3 Oblivion: 3 Paradox: 3 Quixotic: 1 Serendipity: 3 Sublime: 3 Zenith: 3

[simonw]

Randomness is difficult. I wouldn't expect any LLM to be able to reliably produce random anything, except in the cases where they have access to tools (ChatGPT Code Interpreter could use Python's random.random() for example).

[freejazz]

Okay, but repeating words?

[Terr_]

Makes sense to me: It's not understanding and complying with your request, it's doing statistics to guess what kind of follow-up would be most common.

[MattGaiser]

Are you using regular or pro? As pro has no issues with this task.

[hx2a]

Regular user. For useful tasks, how much of a difference does the Pro plan make?

[MattGaiser]

Nowhere near as good as ChatGPT 4o or Claude (in not one case have I had it outperform those other two), but at least it can do math and data science correctly most of the time compared to the regular model.

I use it as a secondary when the other two are chewing on other tasks already.

I only own it as I am an outrageously heavy consumer of LLMs for all sorts of little projects at once and they all seem to pause one window if you use another.