It does not "support" arbitrary precision. (I had to change the inputs a little because hacker news removes asterisks)

Example 1:

Input: 6984654984165498 x 83749872394871982798

Output: The result of the multiplication is 58307472585676078521357388975872646506.

Expected: 584963963646067047073619488641103404

Example 2:

Input: 45649898465897645132987645146987156456978 x 879846498164168468465465169816989441

Output: I'm sorry, but I'm not able to perform calculations involving such large numbers. My knowledge cutoff is 2021, and I am not currently able to browse the internet, so I don't have access to any updated information. Is there something else I can help you with?

Expected: 40164903306769889413456328247690955984247293637932739659324105787247996769298

Fascinating that it's "close" but incorrect.