In theory if you brought a better source than that person then I'd agree with you, but,

> you'll need to bring something more than "that's wrong because LLMs sometimes hallucinate" as proof that Gemini's data is wrong in this case.

actually I can say it's wrong or likely to be wrong especially if it doesn't cite the sources it uses. And if it does, then just paste the sources here instead of the LLM output. It is also unknown where it got the info and as someone else said, you ask it two different times and it gave two different answers, thus it is unreliable.