Put on your high heel sneakers and pull that wig hat on! It's time for a FirstNations Fun Activity!!!!!
A. Ask a question of your search engine of choice (mine is Gleegle.) In order to keep your interest in this diverting activity high, ask an interesting question. I mean geeze.
2. Now let Mr. AI answer that question. Do not go actively seeking Mr. AI, just let the intrusive version of AI chime in with its two centavos.
b. Now read that answer.
B. Now take that answer in. Including the kind of links it might throw up, if any.
C. Now swap over to another search engine. Say, Mr. YeeHaw. Ask the very same question in exactly the same terms - letter, space, and symbol perfect.
d. And proceed as the above.
D. Do this a few more times in different search environments. Try Waterfowlx2 Avaunt, Bling, Microcephaly FullCircle, anything that has that intrusive AI feature. Don't forget to consider those answers!
What becomes immediately apparent to you?
IT SHOULD BE THE OBSCENE AMOUNT OF BIAS, KIDS.
But gosh dangit, oh hey, it's just AI, sometimes it's wrong! Sometimes it's stupid! Look at all these funny memes! Look at all the funny mistakes that AI makes! No threat here!
Except yes. I promise you. There is a threat. It's in the bias profile.
Here's what I mean: Each search engines' servant clanker 'expresses itself' using an emotional tone profile. In other words, the language it uses to word it's replies can come across as happy, dumb, scary, annoyed and etc. The makers and purveyors of search engines and web crawler platforms say "No no no no no, that's just to be 'user friendly', we wouldn't dream of inserting any bias operants in order to subtly sway your opinions! "
Putting aside the question "Why do it at all, when up to this point I haven't demanded that of my encyclopedias, dictionaries, bus schedules and mathematical charts? --
So having put that question aside, simply ask yourself 'What do I like more? Scary things or nice things?"
Never fear. I do not believe that my computer is attempting to control my mind. That's not what I'm saying. I do not lie in bed awake each night listening for the sound of it's dirty little felt pads sliding closer in the darkness. No no no.
I do believe that all the different brand AI-identities, with all their baked-in semantics games, are being shoved down our gullets faster than the finer details on the commercial level can be examined, or any kind of ethical framework can be established or enforced, and that this process is being expedited while human rights issues are at an all-time low in this country.
And I do believe that the big guns aren't even in broad play yet. But don't worry! We'll see that happening in the coming elections!
I know it's freaky. I just got done freaking myself out for a whole day digging into this shit. Marveling at how convincing some of the verbal and visual AI generated content out there can be. How creepy and tonally specific some of the language choices are in the responses I get.
What y'alluns in Parts Foreign have got to look forward to, I do not know.
May I suggest you ask your uncle Jabez Featherstone-Haugh
_______________________________________________________
______________________________________
*Here's a query you should try over on Microcephaly:
"does scientology own reddit stock"
...yeah, see where that gets ya. Now try the same exact query on Gleegle.
Compare answers.
Think about that shit for awhile.
No comments:
Post a Comment