Yup, any time someone can massage input, the output can be trashed. Statisticians and pollsters know this and rely on it. Beware of geeks bearing grifts.
And that's the sum total. You have larger more complex calculations but it's still the same old adding machine at the base of it. An extremely useful adding machine, I'm not knocking it. I've made AI work for me a couple times. But it's 1s and 0s through the registers just like we learned in A+.
I've gotten grief for refusing to fully trust AI, but it's Friday. I might as well open myself up for more grief. 🤷♂️😁
To me, relying on AI is like relying on a neighbor's screwdriver that they may have ground down for a specific purpose, rendering it useless for yours. Or they may have moved, leaving you without a screwdriver when you need it. Or they might even start asking you for a cash deposit every time you show up asking to borrow a tool that no longer works that well.
I keep saying it over and over. It may be a useful tool in certain circumstances. Treat it like the intern. Give it very specific instructions, check its work, and don't trust it with anything of real importance. I use it to populate data into spreadsheets that are formatted into something I can actually use. It still won't bring me coffee though.
It's a (somewhat) old scifi trope. Limits to development because the AI goes insane after a while. Surprise, surprise. Maybe it's true. Although it probably only started off as a convenient plot device.
This is the case with all AI. It's why 'AI Fighter Jets' will always end up suborned. Hell, Iran did it over a decade ago with secret CIA 'AI' Drones, stealing one of them.
I continue to use AI to generate occasional art (that I ALWAYS have to doctor in PS), but have resisted using it to generate text projects. I know people have been using it for analysis, plot generation, idea generation. Sorry, I need to use my brain for that or I'm going to get duller than I already am. It's very tempting, because I struggle with plotting, but I shall continue to resist temptation. I've been annoyed with autocorrect/complete for years, with its utterly nonsense and tone-incorrect suggestions (on my phone, on discord, on wherever). I had a friend publish some non-fiction with an indie publisher who promised an edit of the work, ran it through some AI garbage machine, and printed a book full of typos, bad grammar, and just overall bad formatting.
Wow, the part about only 250 specially crafted documents being enough really stood out to me. You're so right, this is a massive problem! As an AI enthusiast, it makes you seriously question the scalability of proper data validation for these huge training sets. Such an insightful and truely thought-provoking read, thank you!
Didn't someone once mention something about "garbage in, garbage out"?.....
Yup, any time someone can massage input, the output can be trashed. Statisticians and pollsters know this and rely on it. Beware of geeks bearing grifts.
And that's the sum total. You have larger more complex calculations but it's still the same old adding machine at the base of it. An extremely useful adding machine, I'm not knocking it. I've made AI work for me a couple times. But it's 1s and 0s through the registers just like we learned in A+.
I've gotten grief for refusing to fully trust AI, but it's Friday. I might as well open myself up for more grief. 🤷♂️😁
To me, relying on AI is like relying on a neighbor's screwdriver that they may have ground down for a specific purpose, rendering it useless for yours. Or they may have moved, leaving you without a screwdriver when you need it. Or they might even start asking you for a cash deposit every time you show up asking to borrow a tool that no longer works that well.
I keep saying it over and over. It may be a useful tool in certain circumstances. Treat it like the intern. Give it very specific instructions, check its work, and don't trust it with anything of real importance. I use it to populate data into spreadsheets that are formatted into something I can actually use. It still won't bring me coffee though.
It's a (somewhat) old scifi trope. Limits to development because the AI goes insane after a while. Surprise, surprise. Maybe it's true. Although it probably only started off as a convenient plot device.
This is the case with all AI. It's why 'AI Fighter Jets' will always end up suborned. Hell, Iran did it over a decade ago with secret CIA 'AI' Drones, stealing one of them.
AI is beyond stupid.
Oh, I’d forgotten about ‘that’ one…
I continue to use AI to generate occasional art (that I ALWAYS have to doctor in PS), but have resisted using it to generate text projects. I know people have been using it for analysis, plot generation, idea generation. Sorry, I need to use my brain for that or I'm going to get duller than I already am. It's very tempting, because I struggle with plotting, but I shall continue to resist temptation. I've been annoyed with autocorrect/complete for years, with its utterly nonsense and tone-incorrect suggestions (on my phone, on discord, on wherever). I had a friend publish some non-fiction with an indie publisher who promised an edit of the work, ran it through some AI garbage machine, and printed a book full of typos, bad grammar, and just overall bad formatting.
Wow, the part about only 250 specially crafted documents being enough really stood out to me. You're so right, this is a massive problem! As an AI enthusiast, it makes you seriously question the scalability of proper data validation for these huge training sets. Such an insightful and truely thought-provoking read, thank you!