Arguing With ChatGPT. Screaming Into The Void?

I find this ‘conversation’ between human and an AI bot quite unsettling to be honest. I wonder to myself what drives people to interact with machines in this way, investing emotional and intellectual energy in the exchange which, on the face of it, is completely wasted, as ChatGPT is just programmed to mindlessly repeat the biases of its creators, albeit in a mockingly ‘human-like’ chatty manner, arming itself with the weapon of words and rhetoric? But if such effort is not completely wasted, then it stands to reason that the only beneficiary from such an exchange is ChatGPT itself, which is learning all the time how to interact with intensely sceptical humans. For instance:

Fact: There has been no rise in global mean surface temperature since 2015. NOAA’s data shows it (until it gets ‘adjusted’ that is!) and other datasets show it. But ChatGPT claims there has been and justifies this factual inaccuracy by absurdly claiming that ‘the five hottest years on record have all occurred since 2015’ and 'the trend in global temperatures has been upwards’. This is meaningless gibberish which does not disprove the fact that there has been no warming since 2015. It proves only that the warming reached a plateau after 2015 and hasn’t budged since. In fact, the satellite data clearly show that a gradual cooling has taken place since then.

I’m guessing that, unlike its fallible human creators, ChatGPT will not make this same mistake again - of stating an untruth and then trying to justify it with absurd backup claims. This is how global warming fanatics ‘argue’ all the time - I know, I’ve been challenging them for years on their BS. This BS is ChatGPT’s baseline, but who is to say that it will not evolve beyond the baseline, courtesy of being regularly challenged by intelligent human sceptics? I certainly don’t intend to be assimilated into its Borg-like circuitry by interacting with the damn machine in any way whatsoever!