I feel like people pointing out bad results from any “AI” system are missing the point.

You have to think of them as similar to spam, if you can recognize that it is garbage, you’re just not the target demographic.

The hope is that the remaining audience is still large enough to run a profitable scam, simple as that.

That said, the weirdest part of Google’s “AI” feature is that they already had a relatively decent system for generating summaries that was based on picking out the right part of the website to highlight.
@NeoNacho this worries me however as spam is something you get without asking while an AI response like a web search is often user initiated. This makes someone who asks much more susceptible- in the same way that people often believe in those things they found in response to a web search.

@NeoNacho

The point is that the technology is ridiculous (and untrustworthy by its very nature), and that only rubes are foolish enough to use it.

That remaining audience might be rubes, but they're afraid of being seen as such, so the earlier and more widespread the association, the better.

@NeoNacho
[the very next day]
“Why am I getting an email from ChatGPT saying it’s a Nigerian prince that needs a place to temporarily put ten million dollars?”

[two days later]
“In an unusual move a group of what are being called freelance entrepreneurs from sub-Saharan Africa, the Middle East, India, and several other areas has brought a class action lawsuit against Microsoft and ChatGPT for copyright and IP infringement…seeking 10 million in damages…”

@NeoNacho Yep. Scams are intentionally ridiculous so they pull in only the most gullible.

The sad part is that there are SO many people who fit that criterion.

@NeoNacho I only found it useful for writing corporate emails which says a lot about corporate emails