From Newsgroup: comp.misc

Trump’s “Truth Social” network is offering its own AI chatbot,
specially trained on carefully-chosen Politically Correct™ information sources to ensure that the “facts” that it offers will not contradict official doctrine as subscribed to by those latter-day John Birch
Society refugees that make up his constituency.

Unfortunately, the chatbot has somehow nevertheless become
contaminated with inconvenient “woke” truths like: tariffs are a tax
on Americans, the 2020 Presidential election wasn’t stolen, the Trump family’s cryptocurrency investments do pose a conflict of interest,
and crime is not “totally out of control”. Oh, and Barack Obama is generally regarded the highest in favourability by the general public
among living US Presidents <https://www.nzherald.co.nz/world/too-woke-for-them-trumps-truth-social-ai-challenges-his-claims-on-key-issues/PKHOB5OC2FEJNEVLUNV52EX5DM/>.

How are they going to de-woke the AI? Purge its training and start
again? Which of the original sources did the prohibited information
come from? What if it deduced those truths from a combination of all
those sources? (Because reality does work that way, you know.)

When the Washington Post asked what was being done about this, the
reply was almost hilariously predictable:

The White House declined to comment. Trump Media spokeswoman
Shannon Devine said: “With transparently asinine stories like
this, Washington Post reporters indict themselves as irrelevant
partisan hacks who will probably soon join the growing exodus of
left-wing shills from the paper”.

So the Trump Faithful are supposed to simply ignore the subversive
answers? Because recognizing their existence, on their Great Leader’s
own social network, is itself a surrender to the “woke” agenda?
--- Synchronet 3.21a-Linux NewsLink 1.2