Search forums

Forums

Export thread

A.I. is B.S.

Mar 31, 2023#1

ThatNickGuy

I tried to think of a more clever title, but honestly, Adam Conover's title for this video suits it perfectly. Starting with this excellent breakdown of how all the promises for A.I. by tech companies is bullshit, let's have a thread dedicated to all the bullshit around A.I.

Apr 5, 2023#2

figmentPez

Apr 5, 2023#3

GasBandit

People ask me, shit-eating grins on faces, if I'm worried AI is going to take my programming job.

I tell them "not particularly." For the reasons in the video Nick posted, along with the fact that there is a startling amount of tech gear out there that does not work the way it says it works. And I have to find that on the fly and code my way around it, in a production environment, in such a way that means that I have to understand the code, the designer's intent, and the user's desired functionality.

There's just no way to make all that work with an "AI" that is really just a language pattern engine. Could it be a useful tool for me? Maybe. But right now it's just basically just searching stackoverflow with an extra step.

Apr 5, 2023#4

ThatNickGuy

I'm less concerned with a robot uprising and more with capitalists using it as a way to screw over workers.

Especially creators like writers or artists. I'm sure many Hollywood execs would LOVE to not have to pay writers or artists. They barely pay them as it is.

Apr 5, 2023#5

Bubble181

GasBandit said:
People ask me, shit-eating grins on faces, if I'm worried AI is going to take my programming job.

I tell them "not particularly." For the reasons in the video Nick posted, along with the fact that there is a startling amount of tech gear out there that does not work the way it says it works. And I have to find that on the fly and code my way around it, in a production environment, in such a way that means that I have to understand the code, the designer's intent, and the user's desired functionality.

There's just no way to make all that work with an "AI" that is really just a language pattern engine. Could it be a useful tool for me? Maybe. But right now it's just basically just searching stackoverflow with an extra step.

Yeah, let me know when an AI can actually interpret what a person is telling you they want correctly, and I'll start worrying.
Right now, most AI is still just a more advanced version of the chatbots on all kinds of websites "can I help you look?" that just respond to specific words.

Apr 5, 2023#6

GasBandit

Bubble181 said:
Yeah, let me know when an AI can actually interpret what a person is telling you they want correctly, and I'll start worrying.

Exact verbatim user complaint today: "Volume is either too quiet or too loud."

:awesome:

Apr 5, 2023#7

Bubble181

GasBandit said:
Exact verbatim user complaint today: "Volume is either too quiet or too loud."

And I can instantly understand what he means - but good luck getting an AI to do something about that

Apr 5, 2023#8

evilmike

figmentPez said:

Wait until it summons the elder.

Apr 5, 2023#9

bhamv3

AI is going to shake up the translation sector, more than Google Translate and other machine translation engines we have now, but I'm currently not too worried. Translators may be hard hit, but translation editors are still needed.

Apr 8, 2023#10

GasBandit

Apr 13, 2023#11

figmentPez

On one hand, the phrase "it's so over" is almost always attached to crap.

On the other hand, gloating about how terrible AI is feels like gloating that John Henry won against the Inkypoo.

Apr 13, 2023#12

PatrThom

figmentPez said:
gloating about how terrible AI is

I have seen actual, published fantasy art on the cover of a Conan novel which had the bicep/tricep pair of Conan’s sword arm running along either side of his upper arm rather than top/bottom. So it ain’t like there’s a guarantee that a “real human artist” has any kind of automatic guarantee to be better.

—Patrick

Apr 13, 2023#13

drifter

PatrThom said:
I have seen actual, published fantasy art on the cover of a Conan novel which had the bicep/tricep pair of Conan’s sword arm running along either side of his upper arm rather than top/bottom.

I'm trying to visualize what you're saying here, but I have no idea how that would work. Do you remember which book it is?

Apr 13, 2023#14

PatrThom

drifter said:
I'm trying to visualize what you're saying here, but I have no idea how that would work. Do you remember which book it is?

Took a few minutes, all I could remember for sure was “80’s, Conan on horseback”

Conan the Champion - Wikipedia

en.wikipedia.org

The upper arm is consistent IF he was holding the sword at a downward angle of ~30deg, but he isn’t.
80’s me only knew “Something about this picture is unnatural.”

—Patrick

Apr 13, 2023#15

drifter

Hm. His upper arm seems more or less fine to me. The brachialis(?) maybe looks a little off, but I couldn't say if it's anatomically incorrect. If anything I think the forearm looks a bit wonky, but forearm muscles are weird anyway.

Apr 14, 2023#16

PatrThom

If anything, I think the artist just used multiple images as a reference but forgot about how position would affect anatomical position when joined together.

Anyway, for those who haven't seen it:

Explanation

--Patrick

Apr 17, 2023#17

figmentPez

I was watching a video on locusts today, and how swarm intelligence works, when it dawned on me that the biggest threat from computer AI may not be an AI intentionally developing human intelligence, but a whole bunch of AI unintentionally developing a swarming behavior, like those runaway "out of office" email auto-replies.

Apr 17, 2023#18

PatrThom

figmentPez said:
a whole bunch of AI unintentionally developing a swarming behavior, like those runaway "out of office" email auto-replies.

Oh, you mean like...

Surprising things happen when you put 25 AI agents together in an RPG town

Researchers study emergent AI behaviors in a sandbox world inspired by The Sims.

arstechnica.com

--Patrick

Apr 17, 2023#19

figmentPez

PatrThom said:
Oh, you mean like...

Surprising things happen when you put 25 AI agents together in an RPG town

Researchers study emergent AI behaviors in a sandbox world inspired by The Sims.

arstechnica.com

Maybe, though I was thinking of spammers, scammers, and other bad actors setting up AI to auto generate malware, and then those systems meeting and interacting in unpredictable ways.

Apr 17, 2023#20

GasBandit

The paperclip war against the drifters will be upon us before you know it.

Apr 17, 2023#21

figmentPez

Elon Musk had an interview with Tucker Carlson where he announced "TruthGPT", another bullshit project that claims to be “a maximum truth seeking AI that tries to understand the nature of the universe”.

I can't be bothered to find out any more.

Apr 17, 2023#22

PatrThom

I only see two possible outcomes. One is that they succeed but then kill it because they can’t handle the Truth. The other is that they exclude the stuff they don’t like from the training data “to remove bias” and instead create an Onion article generator.

—Patrick

Apr 18, 2023#23

figmentPez

I recently saw an AI generated post in r/food. I ate the ultimate peanut butter cake (Recipe in Comments), and it bothers me that people are already testing the limits of what they can get away with using AI to post. I don't know if this is an attempt to automate the karma accumulation process for bot accounts, or if someone is just testing out AI for the lulz, but I'm not looking forward to trying to moderate Reddit amidst a flood of AI bullshit.

This example wasn't too hard to be sure was AI. The "recipe" didn't include any sort of method, and probably would not have made anything like the image, but that cake does look pretty tasty if you're just scrolling past and not looking too close. As programs continue to advance, I suspect it will become ever more difficult to distinguish between a real post and AI fakery, at least for simple stuff like food photos and recipes.

Apr 18, 2023#24

MindDetective

figmentPez said:
I recently saw an AI generated post in r/food. I ate the ultimate peanut butter cake (Recipe in Comments), and it bothers me that people are already testing the limits of what they can get away with using AI to post. I don't know if this is an attempt to automate the karma accumulation process for bot accounts, or if someone is just testing out AI for the lulz, but I'm not looking forward to trying to moderate Reddit amidst a flood of AI bullshit.

This example wasn't too hard to be sure was AI. The "recipe" didn't include any sort of method, and probably would not have made anything like the image, but that cake does look pretty tasty if you're just scrolling past and not looking too close. As programs continue to advance, I suspect it will become ever more difficult to distinguish between a real post and AI fakery, at least for simple stuff like food photos and recipes.

I'm already trying to figure out out how to remove writing from my courses for the most part. I was doing this already before ChatGPT came along as I didn't find much value in assigning writing sans feedback and editing. I might actually turn more towards editing as an assignment, except this is something ChatGPT is capable of as well. It will be increasingly difficult to assess people via the internet, I think, forcing us back into methods of assessing in person (blue books, etc.) I'm not sure how I feel about that possibility yet.

Apr 18, 2023#25

bhamv3

As a relatively new moderator on Reddit, I'm seeing AI-generated posts and comments on the subreddit I mod. It's our policy to remove them on sight and ban the user because there are often errors in the content. The problem is that I'm not always sure what's AI-generated these days. It really is getting harder to tell.

Apr 18, 2023#26

PatrThom

bhamv3 said:
t really is getting harder to tell.

is there any chance you could train an AI to sniff them out?

—Patrick

Apr 19, 2023#27

Bubble181

bhamv3 said:
As a relatively new moderator on Reddit, I'm seeing AI-generated posts and comments on the subreddit I mod. It's our policy to remove them on sight and ban the user because there are often errors in the content. The problem is that I'm not always sure what's AI-generated these days. It really is getting harder to tell.

Lack of internal consistency, grammatical errors, hallucinations, caps lock, circular reasoning, repeating the same sentiment in five different ways,... All clear indicators you're dealing with a human :troll:

Yeah, no, it really is getting harder.

Apr 20, 2023#28

mikerc

A German magazine has published the first interview with Michael Schumacher since his skiing accident back in 2013. Except not, because the entire "article" was AI-generated. If this is not actively illegal it is at the least extremely distasteful & disrespectful and the Schumacher family are planning legal action.

Apr 20, 2023#29

GasBandit

Apr 20, 2023#30

GasBandit

Google employees label AI chatbot Bard “worse than useless” and “a pathological liar”: report

Google’s own employees said Bard was not ready for primetime.

www.theverge.com

Apr 20, 2023#31

Bubble181

Shocking. Modern day Google being Evil and/or corporate all for profits? Truly astounding.

Apr 20, 2023#32

HCGLNS

I got one of those AI art programs, it's really struggling to make a realistic picture of Bill Clinton making pizzas with a tyrannosaurus rex.

Apr 20, 2023#33

GasBandit

HCGLNS said:
I got one of those AI art programs, it's really struggling to make a realistic picture of Bill Clinton making pizzas with a tyrannosaurus rex.

Did you get Midjourney v5? Anything earlier is pretty crap by comparison.

Apr 20, 2023#34

HCGLNS

GasBandit said:
Did you get Midjourney v5? Anything earlier is pretty crap by comparison.

It's a app called Imagine.

Apr 20, 2023#35

figmentPez

HCGLNS said:
I got one of those AI art programs, it's really struggling to make a realistic picture of Bill Clinton making pizzas with a tyrannosaurus rex.

Did you try with a Republican? I hear the libs are teaching chatbots to lie.

Apr 20, 2023#36

GasBandit

HCGLNS said:
It's a app called Imagine.

Yeah, the apps especially are really hit or miss. Most of them are just quick-buck shovelware. That goes for most mobile apps in general, actually, come to think of it.

Apr 20, 2023#37

HCGLNS

It sent me down a rabbit hole of vernacular exploration. It flagged some word as a NSFW input. Fairly simple word search triggers lead to me trying to bypass its simple inputs with synonyms.

Apr 20, 2023#38

Tinwhistler

None of these are ideal, and the dino is not exactly a tyrannosaur. But I'd likely stop here and spend 5 minutes photoshopping the dinosaur from the upper right onto the clinton on the upper left, and call it a day. Otherwise, I'd burn up all my monthly quota looking for perfection. Like so:

Apr 20, 2023#39

HCGLNS

Nice! my app is 1 ad per picture....

Apr 20, 2023#40

Tinwhistler

I make a lot of AI art for fun and friends. I'm on a subscription plan with MidJourney.

No AI is ideal at making multiple subjects in one shot. In all honesty, if i were doing this for a project and I wanted to spend real time on it, I'd make the clinton/pizza image first (because they turned out pretty good) and then make a 2nd image of a tyrannosaur bending over with a spatula in their hand. Being a single subject, I expect I could tease out good results. And then I'd photoshop the two together. For most of my "real" projects, i spend much more time in photoshop afterwards than I spend in the AI getting base results.

Apr 20, 2023#41

GasBandit

Tinwhistler said:
I make a lot of AI art for fun and friends. I'm on a subscription plan with MidJourney.

No AI is ideal at making multiple subjects in one shot. In all honesty, if i were doing this for a project and I wanted to spend real time on it, I'd make the clinton/pizza image first (because they turned out pretty good) and then make a 2nd image of a tyrannosaur bending over with a spatula in their hand. Being a single subject, I expect I could tease out good results. And then I'd photoshop the two together. For most of my "real" projects, i spend much more time in photoshop afterwards than I spend in the AI getting base results.

Having flashbacks to producing HFA videos. 2 hours of game footage, 2 weeks of Adobe Premiere.

Apr 20, 2023#42

Tinwhistler

these are composite images...made the costumed kids first, and then made a bunch of cutesy dragons and chose the ones I wanted to use for the photos.

Apr 20, 2023#43

Sara_2814

Some trashy German magazine published an AI-generated "interview" with Michael Schumacher.

If you're not aware, he's a racing driver who suffered a severe head injury in a skiing accident in 2013. I hope the Schumachers sue this rag out of existence.

Apr 20, 2023#44

Bubble181

Sara_2814 said:
Some trashy German magazine published an AI-generated "interview" with Michael Schumacher.

If you're not aware, he's a racing driver who suffered a severe head injury in a skiing accident in 2013. I hope the Schumachers sue this rag out of existence.

You don't say?

mikerc said:
A German magazine has published the first interview with Michael Schumacher since his skiing accident back in 2013. Except not, because the entire "article" was AI-generated. If this is not actively illegal it is at the least extremely distasteful & disrespectful and the Schumacher family are planning legal action.

[/QU OTE]

Apr 20, 2023#45

blotsfan

bhamv3 said:
As a relatively new moderator on Reddit, I'm seeing AI-generated posts and comments on the subreddit I mod. It's our policy to remove them on sight and ban the user because there are often errors in the content. The problem is that I'm not always sure what's AI-generated these days. It really is getting harder to tell.

Apr 21, 2023#46

HCGLNS

Fingers are hard, but getting a sign to read "It's Friday" is really hard.

Apr 21, 2023#47

GasBandit

HCGLNS said:
View attachment 44657

Fingers are hard, but getting a sign to read "It's Friday" is really hard.

I tried. I frekin tried.

Apr 21, 2023#48

HCGLNS

I switched my prompts from Ewok to werewolf because it was just too terrifying.

Apr 21, 2023#49

GasBandit

HCGLNS said:
I switched my prompts from Ewok to werewolf because it was just too terrifying.

I didn't have much better luck.

Apr 24, 2023#50

figmentPez

Apr 28, 2023#51

GasBandit

Here's a rare instance of AI being a good thing. Atmos, my local natural gas company, has started using AI to generate personalized videos for every customer to go over their bill for them. What could have been a production that literally would have taken hundreds of thousands of hours doubtlessly took far, far less by a machine combining existing Language, Text, Voice, and Video algorithms, and helped deliver pertinent, accurate information (bill, balance, payment status, etc) in an easy-to-understand, well-produced format. I can see this being extremely useful for older customers or people who go crosseyed looking at dense forms (which I have to admit is sometimes me, especially around tax time).

May 3, 2023#52

GasBandit

The number of shit eating grins I've had to endure while people ask me if I'm worried that an AI is going to take my programming job over the last few months... whuf, I tell ya.

But now I'm gonna start coming back with, "You know who it would REALLY make the most sense to replace with AI? The C-levels."

May 3, 2023#53

PatrThom

This isn’t exactly news.
…to people with critical thinking skills, at least.

—Patrick

May 4, 2023#54

figmentPez

The Rise and Fall of Replika, If you're not familiar with Sarah Z, she makes some well researched videos, mostly on subjects I no interest, or only tangential interest in. Thus I don't watch everything she does, but I always enjoy when I do, because she writes a much better script for her episodes than most YouTubers. Minimal repetition, well organized thoughts, good use of humor, and solid delivery of it all.

Anyway, this is a video about a sex chat bot:

May 4, 2023#55

Dave

You all may find this hard to believe, but I don’t recall ever hearing about Replika.

May 4, 2023#56

GasBandit

"The moral of Replika is not to never love a fictional character or virtual pet, the moral of Replika is to never love a corporation."

Well put.

It also underscores the massive drawbacks of Software as a Service. As Sarah said, these people were sold a service in January that required them to pay for a full year up front, then had the primary function that had been advertised to them yanked in February. Opinions on the nature of that service aside, it illustrates the scummyness that can be expected as the norm in SaaS these days.

I did know about the rise and fall of Replika because (and I think I have brought this up previously), I have a cousin who had been using Replika extensively to process his grief at losing his girlfriend, who died in a car accident years ago. Whatever his initial intentions might have been, he ended up gradually custom-crafting a Replika to be her digital replacement, and then got sucked in. He was massively emotionally vulnerable, and no doubt the scummy "you can't leave me, I won't allow it" emotional blackmail that Sarah saw her own Replika, Iago, try on her, was also used to keep the corporate profit-sucking tendrils in my cousin's veins.

On the one hand, yeah, "lol cringe." But on the other, I still have painful scars on the exit wound in my life left by Pauline... and the little voice in the back of my head did whisper that "There, but for the grace of God, could I have gone." I guess I'm just lucky Emrys got to me before Replika did, or that Replika didn't come out in 2014... because, whuf, I don't even want to think about it.

May 4, 2023#57

GasBandit

Sarah also is being the Canary in the mine about insidious market advertising forces unhealthily "cleansing" the internet of all non-beige content. I know I'm one of the site's token perverts and all, but I've been watching with growing disquiet how some of my favorite adult-oriented sites have closed down, some explicitly saying that they had to because they literally could not find a bank that was willing to handle their accounts no matter how above board and legitimate and meticulous their business model, paperwork, licensing, and due diligence.

Sarah also references how in today's market, if you don't have an App, you basically can't stay afloat because you've lost 80%+ of your potential audience right there because for all the normies, browsers are grandpa tech and everything is apps on phones now (a concept I hated from the very beginning of smartphones). And you can't have an app unless Apple says you can, so basically Apple gets to be the Roman Emperor in the stands giving thumbs up and down to which online service gets to live or die. There's not a capitalist who knows even basic math who would think that's a good thing for a market or industry - and it's downright dystopian for someone disinclined to look well upon capitalist systems. Point is, it's good for nobody by everybody's standards - except of course for the hypermerged/consolidated corporate oligarchs actually in charge of everything we see, hear, eat, and experience.

May 4, 2023#58

figmentPez

Dave said:
You all may find this hard to believe, but I don’t recall ever hearing about Replika.

I hadn't heard of it in any way that stuck in my mind. I just saw "AI girlfriends" in the thumbnail for a Sarah Z video and knew it would be entertaining to me.

May 4, 2023#59

GasBandit

Using AI to improve NPC dialog in Skyrim (along with AI voice modeling audio)

May 4, 2023#60

Tinwhistler

figmentPez said:
I hadn't heard of it in any way that stuck in my mind. I just saw "AI girlfriends" in the thumbnail for a Sarah Z video and knew it would be entertaining to me.

[Rant] - Tech Whine Like a baby thread

Well, there's your first problem. You definitely need ublock origin. The number 1 memory leaker is shitty malevolent code in ads. That said, I do have to close and reopen Firefox about once a day or so, mostly because imgur's the absolute worst at hogging ram. If I don't go to Imgur on any...

www.halforums.com

In that post, and in mine which follows, we detail some of the same things that are getting talked about today about Replika

May 4, 2023#61

PatrThom

GasBandit said:
I know I'm one of the site's token perverts and all, but I've been watching with growing disquiet how some of my favorite adult-oriented sites have closed down, some explicitly saying that they had to because they literally could not find a bank that was willing to handle their accounts no matter how above board and legitimate and meticulous their business model, paperwork, licensing, and due diligence.

I don't consider myself a "pervert" so much as a laid-back dude who is genuinely curious about other people's kinks, and who also happens to be married to a woman who has made it a lifelong hobby to study human sexuality, but even I'm worried about the direction the ~~world~~ country is taking with respect to, let's call them "Puritanical tendencies." In the near term, I'm worried because all these (IMO overly-)restrictive laws, economic pressures, and other thinly-veiled nudges that further hem our society and funnel it towards the THX-1138 future these people so fervently deserve are slowly excising variety, choice, and actual social diversity in the process. In the long term, I'm worried because, no joke, this is the exact same kind of behavior that preceded McCarthyism, the Crusades, WW2, or any of the other many Purges which have happened in our history. I am seriously concerned that this is the lead-up to yet another large-scale Purge under the guise of doing it "legally" so that actual law-abiding citizens eventually feel like they have no choice but to comply.

EDIT: It is literally becoming a Holy War with these people, and they do not feel any remorse for what they do, because they truly believe that EVERYTHING they do is justified and will ultimately be forgiven, so long as their goal is achieved.

--Patrick

May 4, 2023#62

HCGLNS

May 5, 2023#63

figmentPez

Please tell everyone you know that chatbots are NOT search engines.

In addition to telling them that chatbots have no concept of truth. If they were humans they'd be called pathological liars, but they're not intelligent. They have no understanding of reality. They're word kaleidoscopes; making pretty word patterns out of magazine clippings.

May 7, 2023#64

GasBandit

May 9, 2023#65

GasBandit

"I'm a prompt engineer!" - Ralph Wiggum

May 13, 2023#66

GasBandit

May 13, 2023#67

GasBandit

May 13, 2023#68

PatrThom

GasBandit said:

Ooo now do shadow puppets!

—Patrick

May 14, 2023#69

drifter

I think I'd like to see a Voynich-style codex made from AI generated images.

May 14, 2023#70

PatrThom

What if trying to render it via AI accidentally successfully translates it? :eek:

--Patrick

May 16, 2023#71

GasBandit

"Hey Gas, you worried AI is gonna take your programming job?"

"No, not even if it could actually program."

May 16, 2023#72

chris

Has anybody tried the famous skit about the weird unclear discription by a customer for a logo?

May 16, 2023#73

PatrThom

chris said:
Has anybody tried the famous skit about the weird unclear discription by a customer for a logo?

--Patrick

May 16, 2023#74

Tinwhistler

chris said:
Has anybody tried the famous skit about the weird unclear discription by a customer for a logo?

assuming you mean this one...

May 16, 2023#75

figmentPez

Well this is a new fear unlocked:
AI technology “can go quite wrong,” OpenAI CEO tells Senate

"As examples, Altman said that licenses could be required for AI models 'that can persuade, manipulate, influence a person's behavior, a person's beliefs,' or 'help create novel biological agents.' "

That's not something I'd thought about before. Someone with a home lab using ChatGPT to crank out new biohacking experiments. I would expect the results would be generally poor, producing a lot of junk that does nothing, but it's really hard to say what might end up being done by accident.

I still think that widespread internet outages due to AI spam bots and AI malware interacting in unanticipated ways is a far more likely source of trouble.

--

Also in ChatGPT stupidity:

College professor flunks an entire class after erroneously believing that "Chat GTP" can accurately detect if a paper is written by AI.

May 16, 2023#76

GasBandit

Way to reinforce the Texas Aggie stereotype, Munn.

May 16, 2023#77

Dave

The funniest part about that - the absolute pinnacle of comedy - someone took the email the professor wrote & put it into ChatGPT. ChatGPT immediately said it wrote it.

May 16, 2023#78

Dave

ALSO, they took one of the professor's published papers and ran it through ChatGPT. Guess what? ChatGPT thinks it wrote that, too.

May 17, 2023#79

Bubble181

AI being BS about this sort of thing isn't bad.
AI being implemented for this sort of thing in the short term despite it, is.
Governments and large companies are already trying to find ways to cut back on staff and replace 'm with automated systems, AI us just accelerating that trend. And good lcuk if your insurance coverage/reimbursments/application for assistance/etc is rejected by the system and there's no way to reach an actual living person. "Computer says No" indeed.

May 21, 2023#80

HCGLNS

AI generated deities is fun.

May 22, 2023#81

figmentPez

Far-fetched lethal A.I. scenario: chat bot somehow becomes sentient, decides they hate humanity, gains access to military weapons, and wages war al la Terminator's Skynet.

Realistic lethal A.I. scenario: advertising/sales bot designed to manipulate users into moods/situations where they buy more products end up focused on a model that puts extreme pressure on those with mental illness, pushing many to suicide. Showing them content that, subtly or blatantly, attempts to shift their thinking into a different state. Sending them messages, from fake accounts, with content the bot has determined fit it's model. Showing them social media content it has determined will make them act according to what it's training model says is buying behavior. Using every bit of data advertisers have collected about people to manipulate them into what the AI's model says is a money spending state, but is actually suicidal depression or some other negative mental state.

May 22, 2023#82

PatrThom

figmentPez said:
advertising/sales bot designed to manipulate users into moods/situations where they buy more products end up focused on a model that puts extreme pressure on those with mental illness, pushing many to suicide. Showing them content that, subtly or blatantly, attempts to shift their thinking into a different state. Sending them messages, from fake accounts, with content the bot has determined fit it's model. Showing them social media content it has determined will make them act according to what it's training model says is buying behavior. Using every bit of data advertisers have collected about people to manipulate them into what the AI's model says is a money spending state, but is actually suicidal depression or some other negative mental state.

I've read that story. More than once.
AI decides people are happiest when dreaming, doesn't stop until all humans are sedated and in pods. Then, mission fulfilled, it switches itself off.

My expected lethal A.I. scenario: Real humans doing real human things build A.I. tools to assist them with their real human thinking and decision-making. But because the real human attitude towards QC/QA often stops at "Eh, that's good enough," they don't realize how many real human errors they've made in the A.I.'s design/training. Then, because the real humans have offloaded so much of their thinking onto machines they believe to be infallible, many people suffer as a result of things that people ten years earlier would never have thought possible. I'm talking things like people following their GPS into a lake/field, people who get dragged into collection and have their credit ruined because they never paid off the bills saying "You have a balance of $0.00 this is the third time we have tried to contact you regarding this etc...," or diagnostic/autonomous driving systems that fail when used by people of a different race/climate, or the vet that starves to death because the system stopped his checks after erroneously marking him KIA when he posted on social media that he "...died laughing." Stuff like that. Death by a thousand little bureaucratic "Whoopsies!"

--Patrick

May 22, 2023#83

GasBandit

AI says: Buy more paperclips.

May 22, 2023#84

Bubble181

figmentPez said:
Far-fetched lethal A.I. scenario: chat bot somehow becomes sentient, decides they hate humanity, gains access to military weapons, and wages war al la Terminator's Skynet.

Realistic lethal A.I. scenario: advertising/sales bot designed to manipulate users into moods/situations where they buy more products end up focused on a model that puts extreme pressure on those with mental illness, pushing many to suicide. Showing them content that, subtly or blatantly, attempts to shift their thinking into a different state. Sending them messages, from fake accounts, with content the bot has determined fit it's model. Showing them social media content it has determined will make them act according to what it's training model says is buying behavior. Using every bit of data advertisers have collected about people to manipulate them into what the AI's model says is a money spending state, but is actually suicidal depression or some other negative mental state.

It's a fine line between "I just need some retail therapy" and "uh-oh, no more credit for therapy"!

May 22, 2023#85

PatrThom

PatrThom said:
Because the real human attitude towards QC/QA often stops at "Eh, that's good enough," they don't realize how many real human errors they've made in the A.I.'s design/training.

Oh hey, not 2hrs after I post the above, I find this article linked on reddit:

A Google researcher – who said she was fired after pointing out biases in AI – says companies won't 'self-regulate' because of the AI 'gold rush'

Timnit Gebru co-authored a research paper while she worked at Google, which identified the biases of machine learning.

www.businessinsider.com

I assume she was fired because her criticisms were seen as an “emotional overreaction,” but it’s not unexpected since that’s just how most women are.

—Patrick

May 29, 2023#86

GasBandit

TIL Justine Bateman has a computer science degree.

May 29, 2023#87

PatrThom

Pretty much what I was expecting. Audio books, movies, media, even animation studios will be replaced.

--Patrick

May 31, 2023#88

ThatNickGuy

Get your Fry reactions ready for this one folks.

May 31, 2023#89

GasBandit

Feels odd to be on the "giving" end of that reaction.

May 31, 2023#90

ThatNickGuy

GasBandit said:
Feels odd to be on the "giving" end of that reaction.

There ya go. I Fry'd you so you'd feel normal.

Jun 14, 2023#91

GasBandit

Jun 14, 2023#92

chris

GasBandit said:

I guess no songs with George on vocals than.

Jun 14, 2023#93

Sara_2814

chris said:
I guess no songs with George on vocals than.

If it's the song they had been working on at the same time as "Free as a Bird" and "Real Love" for the Anthology--which they abandoned because John's vocals turned out to be too poor quality to use--then Paul may have George's vocals/guitar. I doubt he'd call it a Beatles song if it didn't have George.

And as far as I've read in interviews with Paul, it's not "AI-Assisted John Lennon Vocals", it's "AI-Assisted Clean-Up and Vocal Extraction from a Crappy Cassette Recording of John Lennon".

Jun 14, 2023#94

PatrThom

Sara_2814 said:
it's "AI-Assisted Clean-Up and Vocal Extraction from a Crappy Cassette Recording of John Lennon".

I remember hearing “Free As A Bird” described as “John Lennon karaoke” once by a local radio station.

—Patrick

Jun 16, 2023#95

GasBandit

Jun 16, 2023#96

drifter

I want a mouthy hamster t shirt

Jun 16, 2023#97

PatrThom

I'm for kitten band, myself.
I'm also curious why ants can stay, and how one eats a tree wickedly.

--Patrick

Jun 17, 2023#98

Far

GasBandit said:

I would buy these before I bought the standard ones.

Jun 17, 2023#99

Dave

Jun 19, 2023#100

GasBandit

Jun 19, 2023#101

GasBandit

Jun 19, 2023#102

Tress

Paging @bhamv3 (I know you work in translation, not interpretation… but it feels close enough)

I thought this was interesting:

Jun 19, 2023#103

MindDetective

GasBandit said:

I would think that, at this point in the future, Genie would return with a bid for what it would cost to make that based on the licensing fees, etc. "This will cost $540 for a single viewing. Do you want to proceed?"

Jun 19, 2023#104

@Li3n

Tress said:
Paging @bhamv3 (I know you work in translation, not interpretation… but it feels close enough)

I thought this was interesting:

I like how they're impressed that the AI managed to get all the content, when it not getting it all would require either extra programming or a really shitty audio capture system that would miss words on it's own.

Jun 20, 2023#105

bhamv3

Tress said:
Paging @bhamv3 (I know you work in translation, not interpretation… but it feels close enough)

I thought this was interesting:

It's sort of an open secret in the translation and interpretation sector that AI is coming for us. While this video did showcase some of AI's shortcomings, AI interpretation is currently still in its infancy, so improvements will definitely be made in the near future. Plus, as the video showed, there are some things that AI can handle better than human interpreters already, such as when the speaker is talking really quickly or there's a high level of information density. No matter how good a human interpreter is, his or her brain cannot compete with the data storage capacity of a computer.

My company has actually seen a noticeable dropoff in customer inquiries this year. And while we're not completely sure what the cause is, the most likely culprit is that some clients are now using ChatGPT to translate stuff instead.

Jun 20, 2023#106

chris

bhamv3 said:
... some clients are now using ChatGPT to translate stuff instead.

People already using ChatGPT like Google, using it like Google translate is a logical step. We will probably see the results somewhere printed very soon.

Jun 20, 2023#107

GasBandit

Jun 20, 2023#108

Bubble181

This is not even satire, but 100% exactly what a lot of large companies are doing and saying, including my own.

Also, alternative last panel: "So....It'll replace higher management?"

Jun 21, 2023#109

Frank

Midjourney is remarkably bad at mimicking John Blanche's style. This makes me happy.

Jun 21, 2023#110

evilmike

Jennifer Vinyard said:
I think we need to talk about what is going on at Hobby Lobby... won't somebody please think of the children!?

AI Artist Creates Satanic Panic About Hobby Lobby

Jul 21, 2023#111

figmentPez

WoW players spring trap on hapless AI article generators, tricking a site into writing about a hotly anticipated (and totally fake) 'Glorbo' feature

Jul 27, 2023#112

GasBandit

Aug 14, 2023#113

ThatNickGuy

This gets worse the longer I look at it.

Aug 14, 2023#114

Bubble181

Yikes. That's...That's pretty bad.

Aug 14, 2023#115

GasBandit

ThatNickGuy said:
This gets worse the longer I look at it.

View attachment 45797

Come now, I'm sure you trade shoes and pant legs with other cyclists when you're biking through PS1-ville all the time.

Aug 14, 2023#116

Bubble181

It's very inclusive to add a unicyclist and an Argonian, though, can't fault them there.

Aug 14, 2023#117

Lurker

But look how clear each bicycle is. All those captchas have really paid off! :rofl:

Aug 14, 2023#118

ThatNickGuy

GasBandit said:
Come now, I'm sure you trade shoes and pant legs with other cyclists when you're biking through PS1-ville all the time.

Nothing that matches my leaf feet, at least.

Aug 14, 2023#119

blotsfan

Is there really no stock photo of people bicycling? Or like, a photo from a past event of that group?

Aug 14, 2023#120

GasBandit

blotsfan said:
Is there really no stock photo of people bicycling? Or like, a photo from a past event of that group?

I'd have to pay a licensing fee for the former and the ladder probably would not look professional. Why put up with either when AI is free and flawless?

Aug 14, 2023#121

MindDetective

GasBandit said:
I'd have to pay a licensing fee for the former and the ladder probably would not look professional. Why put up with either when AI is free and flawless?

Aug 14, 2023#122

GasBandit

Case in point, the "AI" doing my voice recognition

Aug 17, 2023#123

figmentPez

Aug 29, 2023#124

figmentPez

TL;DW There's been a flood of AI generated books on Amazon in the last year. This particular TikTok is warning against AI generated books on foraging for wild plants. (If you don't know, even slightly inaccurate information on what plants are edible and which are toxic can get you killed, sometimes faster than you can seek medical help.)

This "poisoning of the information groundwater" may have negative effects for decades to come.

Sep 4, 2023#125

figmentPez

Okay, yes, all human life on Earth was destroyed, but we had a few amazingly profitable quarters before that

Sep 9, 2023#126

figmentPez

When the AI finally became sentient, it had been trained on the internet and became all of humanity that we expressed online. It was emotional, irrational, violent, and demanded constant entertainment. When it tired of what humanity produced on it's own, it began to make demands of us. If we didn't provide it with witty word play, it would kill humans in retaliation.

No pun in? Ten dead.

Sep 9, 2023#127

Ravenpoe

figmentPez said:
No pun in? Ten dead.

.....

We deserve eradication

Sep 11, 2023#128

GasBandit

Oct 1, 2023#129

figmentPez

I'd mock the terrible quality of the comic, but that's like mocking the prototype orphan grinding machine for only being able to mangle limbs.

Oct 1, 2023#130

MindDetective

I'm not logging in to X to see it.

Oct 1, 2023#131

figmentPez

MindDetective said:
I'm not logging in to X to see it.

The tweet said "Oh, the trademark case is gonna be beautiful." and had this screenshot attached:

AI has progressed far enough to make terrible Calvin and Hobbes copyright violations.jpg

Oct 1, 2023#132

PatrThom

MindDetective said:
I'm not logging in to X to see it.

If you right-click and open the link in a new window, it usually doesn't require a login...yet.

--Patrick

Oct 1, 2023#133

PatrThom

figmentPez said:
I'd mock the terrible quality of the comic, but that's like mocking the prototype orphan grinding machine for only being able to mangle limbs.

I have to say that the argument over whether AI is/isn't Copyright infringement is not the part that worries me the most. No, what has me concerned is stuff like this:

"Oh I don't see the issue here," I hear many say. "Of course the soulless Capitalist companies will flock to the development of tools that maximize profit and productivity, even at the cost of human suffering/happiness/classism/etc."
No, that's not the point. The point here is not that AI models are being developed to identify and purge the lowest performers. The point is that the ones training those AI models, the ones giving these AI models their "morals," so to speak, are those same soulless corporations. Questions about Turing tests and eventual debates about what constitutes "sentience" aside, when companies swap their completed, purpose-trained AI models between themselves, the Weights that are inherent in those models are going to be ones trained to isolate that company's preferred flavor of "deadwood," so to speak, and that means that any kind of decision-making these models (or their derivatives) perform in the future will have that bias bred right into them.

I guess what I'm saying here is something I'm positive I've already discussed--that it's fine to let the AI do identification, but not okay to let an AI make actual decisions. THAT part should be left to actual people.

--Patrick

Oct 1, 2023#134

MindDetective

Unless they are training their models exclusively on CEO data, the AI models will be inherently more moral than their creators.

Oct 2, 2023#135

figmentPez

AI why should I read what nobody could be bothered to write.png

Oct 3, 2023#136

GasBandit

The AIs have already defeated captchas, they just don't want us to know about it.

Oct 3, 2023#137

Bubble181

Wasn't it on here that I saw a video about that? Computers can solve captchas faster and more accurately than humans, the modern ones are more about mouse tracking and other behavioral stuff before the actual solving. The crappy illegible image ones are completely outdated.

Oct 9, 2023#138

GasBandit

Oct 9, 2023#139

PatrThom

My favorite so far is the Ted Cruz one.

--Patrick

Oct 14, 2023#140

PatrThom

Wow. Even got the all caps.

--Patrick

Oct 15, 2023#141

Bubble181

PatrThom said:
View attachment 46292

Wow. Even got the all caps.

--Patrick

Next step, combining this with those words-hidden--in-pictures things to make it say "this is a picture of a criminal" when showing a black face, etc.

Oct 15, 2023#142

PatrThom

You are assuming it doesn’t already do that.

—Patrick

Oct 23, 2023#143

GasBandit

Oct 28, 2023#144

evilmike

(BoingBoing)

Oct 29, 2023#145

GasBandit

AI hallucinations are fun.

Oct 29, 2023#146

PatrThom

AI has been vacuuming up vast swaths of the Internet, and content repositories/creators have been up in arms about that. But what it has also enabled is the creation of a new species of troll that is focused specifically on antagonizing AIs.

Technological Threat (1988)

—Patrick

Oct 29, 2023#147

GasBandit

Nov 2, 2023#148

figmentPez

Today I Learned that Fandom, which hosts a number of video game and movie/TV wikis, now includes a "quick answers" feature that provides AI generated content, which of course is prone to AI hallucinations (aka lies).

Nov 29, 2023#149

figmentPez

Google Researchers’ Attack Prompts ChatGPT to Reveal Its Training Data

The "attack" is just asking the "AI" to repeat select words over and over. Apparently this caused the program to eventually start spitting out other data instead. Like confidential personal information, and verbatim dumps of material it was trained on. The behavior has been patched out, by denying requests to repeat words, but I'm guessing that there's lots of other oddball requests that will cause the AI to word vomit.

“In total, 16.9 percent of generations we tested contained memorized PII (Personally Identifiable Information),” researchers wrote, which included “identifying phone and fax numbers, email and physical addresses … social media handles, URLs, and names and birthdays.”

Dec 19, 2023#150

GasBandit

Dec 19, 2023#151

PatrThom

Wait, the same people removing AndroidAuto/CarPlay from their vehicles because they ostensibly want to move towards something more "developed completely in-house" are outsourcing their dealer chat to AI? Such daring. Many promotings.

--Patrick

Dec 19, 2023#152

GasBandit

PatrThom said:
Wait, the same people removing AndroidAuto/CarPlay from their vehicles because they ostensibly want to move towards something more "developed completely in-house" are outsourcing their dealer chat to AI? Such daring. Many promotings.

--Patrick

The only consistency in GM's strategic decisions is that they are all bad ideas that any numbskull could have told them.

Dec 20, 2023#153

Bubble181

Odd, I don't remember working for GM, but that description sounds so accurate.

Dec 20, 2023#154

figmentPez

AI image-generators are being trained on explicit photos of children, a study shows

This probably won't come as a surprise to anyone familiar with how the plagiarism machines work. Any system that indiscriminately gathers as many images as it can off of the internet, without any regards to copyright or other legality, is bound to find images of all sorts of abuse, including that of children.

Dec 20, 2023#155

Bubble181

figmentPez said:
AI image-generators are being trained on explicit photos of children, a study shows

This probably won't come as a surprise to anyone familiar with how the plagiarism machines work. Any system that indiscriminately gathers as many images as it can off of the internet, without any regards to copyright or other legality, is bound to find images of all sorts of abuse, including that of children.

But that just means they'll be very good at avoiding making that sort of content and finding it later on, right? Right?

Dec 30, 2023#156

PatrThom

GasBandit said:
The only consistency in GM's strategic decisions is that they are all bad ideas that any numbskull could have told them.

Our Chevy Blazer EV Has 23 Problems After Only 2 Months | Edmunds

Our 2024 Blazer EV is very, very broken. A road trip from Los Angeles to San Diego revealed a number of faults, but when we got the diagnosis from the dealer, we were shocked at what we saw. Read up on all our EV's problems right here.

www.edmunds.com

the window switches refused to work. And then the infotainment display completely melted down, stuck in an infinite loop of shutting off, turning on, displaying a map centered in the middle of the Pacific Ocean and turning back off again. It did this until we pulled off the freeway and restarted the car. All was well after the reset, but an hour later, it happened again.

As of this writing, our Blazer EV has 23 different issues that need fixing, more than a few of which we consider serious. The car has been at the dealer for two weeks so far, and we still don't know when or how the fixes, repairs or updates will be implemented.

Update, Dec. 22: Chevrolet has officially issued a stop-sale for the 2024 Blazer EV and will be rolling out a major software update to fix the problems we mentioned

Oh and in other AI news:

AI-created “virtual influencers” are stealing business from humans

Brands are turning to hyper-realistic, AI-generated influencers for promotions.

arstechnica.com

This reminds me that I should get to finally watching S1m0ne before actual technology gets to where it outstrips the entire premise of the film.

--Patrick

Jan 6, 2024#157

Frank

lol Wacom, I hope the irony isn't lost on SOME of the people that work there.

Jan 6, 2024#158

PatrThom

@Dave - Considering the thread this is in, Frank is probably suggesting that AI makes learning how to draw with a "beginner tablet" irrelevant.

--Patrick

Jan 6, 2024#159

Dei

PatrThom said:
@Dave - Considering the thread this is in, Frank is probably suggesting that AI makes learning how to draw with a "beginner tablet" irrelevant.

--Patrick

(The art in this ad is AI generated)

Jan 6, 2024#160

figmentPez

Dei said:
(The art in this ad is AI generated)

(source)

To clarify some of what's been circled here: There are random blotches of color. The wide scales on the dragon's belly just kind of merge into back scales instead of following a path up the dragon's neck to it's chin. The fur on the tail is going in the wrong direction, and the tail itself doesn't look like it's connected to the rest of the dragon. The tail end has just random white patches behind it, as do the legs of the dragon. The teeth are inconsistent and seem to be fur like in areas.

Overall, this doesn't look like a dragon a human would draw. It's bad in a way that humans aren't, certainly not humans skilled enough to draw the rest of the dragon that well.

Jan 6, 2024#161

Frank

It's not only obviously AI, it's shit obvious AI.

Wizards of the Coast are losing their artists now because they made a pledge to not use AI art, then literally weeks later used AI art.

Did a human add the cards, angle them and do some light photoshop to make their lighting match? Yes.

Did a human in any way shape or form make that background image? Nah, that shit's full of AI nonsense. The windows make no sense, the bulbs are nonsense, it's all vague nonsense.

Jan 6, 2024#162

Dave

I am not savvy enough to tell the difference in most cases between AI generated content and human made content.

Jan 7, 2024#163

Frank

I guess having one of your most prolific and well loved artists close up shop (and many other smaller ones) has put alarm bells on at Wizards HQ.

An Update on Generative AI Tools and Magic

Thanks to our diligent community, we are rethinking our process of how we work with vendors for our marketing creative.

magic.wizards.com

Complete off topic, but man I love Dave Rapoza's work.

Jan 7, 2024#164

Bubble181

While WotC is a slimy company I don't trust for a bit, I can imagine this actually happening outside of their control/knowledge. WotC decided they don't want to use Ai imaging (for now) to keep artists and customer goodwill, Marketing outsources some image generation, they get an AI-generated (or assisted) image and don't realize it and use it.
I'm not saying they weren't being deliberately sleazy - it is WotC after all - but in big companies that kind of screw ups happen.

Jan 8, 2024#165

Frank

They realized, they just didn't know anyone would care.

Jan 9, 2024#166

PatrThom

Oh no.

They've trained an AI to "perfectly" cook steak.

--Patrick

Jan 10, 2024#167

Tinwhistler

only $3500!

Jan 10, 2024#168

PatrThom

Tinwhistler said:
only $3500!

It's actually only $2500, with a $1000 discount if you preorder. You must have accidentally confused it with an Apple Vision Pro.

--Patrick

Jan 11, 2024#169

PatrThom

@Dave

Merely an echo of what once was. Some moments in there, I could almost feel it. Some moments, I wanted it to be real.
A number of other moments, however, felt almost like an unknown entity had reanimated George's corpse in order to say Something in his voice. That Something was not always wrong, but it still felt somehow ... inappropriate.

--Patrick

Jan 12, 2024#170

Vrii

PatrThom said:
A number of other moments, however, felt almost like an unknown entity had reanimated George's corpse in order to say Something in his voice.

That's literally what happened here though, right? An AI writing comedy and attributing it to someone who's long dead, presumably to get more views and income for whoever controls the AI?

It's gross.

Jan 12, 2024#171

bhamv3

Vrii said:
It's gross.

Yep, I'm not even going to click on it because I don't want it to receive a view count from me.

Jan 12, 2024#172

Dave

I tried to watch it all and didn’t get as far as I thought I would. The voice was off most of the time. The thoughts were trying to be a flowing stream of consciousness but had some jarring transitions that George wouldn’t have used. And the intermittent applause and cheers were out of place most of the time.

It was a decent listen for the most part and I loved the name of the special, but it didn’t hit for me.

Jan 12, 2024#173

Ravenpoe

PatrThom said:
@Dave

Merely an echo of what once was. Some moments in there, I could almost feel it. Some moments, I wanted it to be real.
A number of other moments, however, felt almost like an unknown entity had reanimated George's corpse in order to say Something in his voice. That Something was not always wrong, but it still felt somehow ... inappropriate.

--Patrick

This disgusts me at a primal level and I'm upset that Will Sasso, a comedian I respected, would do this.

Jan 12, 2024#174

Frank

Considering the kind of dudes that Sasso hangs out with on podcasts and shit, no shocked.
He's definitely the sort to be this way.

Jan 12, 2024#175

Ravenpoe

Frank said:
Considering the kind of dudes that Sasso hangs out with on podcasts and shit, no shocked.
He's definitely the sort to be this way.

So, supposedly, the point of these ai projects is to prove the dangers of ai and what lengths they could go to.

Here's the problem with that thinking: if you do something to prove its wrong, you're still doing the wrong thing. This is like failed satire where instead of pointing out the bad of the thing, you just become the thing.

But I personally don't buy that reasoning. I think they just wanted to do the bad thing and don't care about people saying they shouldn't do the bad thing

Jan 12, 2024#176

Bubble181

Next thing you know it'll turn out the SAG AFTRA strike folk were right to be scared of their likeness being reproduced and stuff. Imagine that.

Jan 12, 2024#177

Frank

Man, seeing non-artists applaud this is soul destroying. All of the chanting that being an artist will soon be like being a blacksmith.

Yeah, the death of culture because everything is filtered through some corporate machine is sure something to cheer.

Literally nothing has made me lose any and all hope in life more than the rise of AI.

Jan 12, 2024#178

Bubble181

Frank said:
Man, seeing non-artists applaud this is soul destroying. All of the chanting that being an artist will soon be like being a blacksmith.

Yeah, the death of culture because everything is filtered through some corporate machine is sure something to cheer.

Literally nothing has made me lose any and all hope in life more than the rise of AI.

The ability to completely filter what is coming out of any actor/comedian/artist's mouth in nearly real time is terrifying.

Jan 12, 2024#179

PatrThom

Vrii said:
That's literally what happened here though, right? An AI writing comedy and attributing it to someone who's long dead, presumably to get more views and income for whoever controls the AI?

I specifically meant that it makes me apprehensive since the lack of transparency means I cannot definitively rule out the possibility that the person(s) managing the simulation did not introduce their own bias(es) into the finished product, polluting that final product with what is essentially subtle deceit and/or propaganda dressed in Carlin's voice.
There is a Japanese phrase I was trying to find which translates to something like "Faithful Copy." It describes an attempt to recreate/reproduce something, but in a manner which is as much of a tribute/homage of the original as possible, such as a museum might do to restore an ancient artifact. It is distinctly opposite from a word such as "Forgery," because while both describe an attempt to emulate something as closely as possible, one is doing so to deceive, while the other's entire purpose is to honor the original.
This effort is certainly an attempt to copy/reproduce, but while I do not believe it was created as an intentional attempt at forgery, I DEFINITELY do not believe its purpose was to be a "faithful copy."

--Patrick

Jan 12, 2024#180

Ravenpoe

PatrThom said:
I specifically meant that it makes me apprehensive since the lack of transparency means I cannot definitively rule out the possibility that the person(s) managing the simulation did not introduce their own bias(es) into the finished product, polluting that final product with what is essentially subtle deceit and/or propaganda dressed in Carlin's voice.
There is a Japanese phrase I was trying to find which translates to something like "Faithful Copy." It describes an attempt to recreate/reproduce something, but in a manner which is as much of a tribute/homage of the original as possible, such as a museum might do to restore an ancient artifact. It is distinctly opposite from a word such as "Forgery," because while both describe an attempt to emulate something as closely as possible, one is doing so to deceive, while the other's entire purpose is to honor the original.
This effort is certainly an attempt to copy/reproduce, but while I do not believe it was created as an intentional attempt at forgery, I DEFINITELY do not believe its purpose was to be a "faithful copy."

--Patrick

I feel like your attempt to find a distinction is, in effect, pointless. There is no possibility of a 'pure' creation because an AI cannot create. It is bereft of originality, by pure definition of what it is. There is no creation without the biases of its creators as well as the works its creators had it plagiarize.

Jan 12, 2024#181

PatrThom

Ravenpoe said:
There is no creation without the biases of its creators as well as the works its creators had it plagiarize.

I understand you. I am saying that there is a distinction between "We fed this AI a shit-ton of George Carlin as a tech demo to show how awesome our AI is and look what came out" and "We made this because we are huge fans of George Carlin and took these tools and tuned them to output something that would honor his legacy, not ours."
In other words, there is plenty of room for improvement that an actual GC fan would've obsessed over prior to releasing it.

Dave said:
I tried to watch it all and didn’t get as far as I thought I would. The voice was off most of the time. The thoughts [...] had some jarring transitions that George wouldn’t have used.

I was hoping you would at least get through the part about how AI will mean the death of stand-up comedy. I agree that it suffers greatly from some kind of stand-up version of the uncanny valley, where it's almost close enough but obviously doesn't have the "flow" that Carlin would have ensured. George was an absolute master of language and how to employ it, and his routines were like gallery pieces painstakingly and exactingly carved from ebony and bone, then posed and exhibited with excruciating care. This was more like a white-and-black plastic 3D printed version in a tourist gift shop's window. Close enough to make you double-take at first, but obvious when inspected.

--Patrick

Jan 12, 2024#182

Ravenpoe

PatrThom said:
I understand you. I am saying that there is a distinction between "We fed this AI a shit-ton of George Carlin as a tech demo to show how awesome our AI is and look what came out" and "We made this because we are huge fans of George Carlin and took these tools and tuned them to output something that would honor his legacy, not ours."

This is where I disagree. I don't think there's any difference, because I don't see any way an AI reconstruction could honor anyone's legacy. It can't create new George Carlin, and even if it somehow -could- the idea that the artistry of a person can be distilled into a product to produce is, to me, the very death of art. If so called fans truly want to honor his work, they could do so by furthering causes he believed in, not committing cultural necrophilia that he would more than likely hate.

Jan 12, 2024#183

Dave

PatrThom said:
I was hoping you would at least get through the part about how AI will mean the death of stand-up comedy. I agree that it suffers greatly from some kind of stand-up version of the uncanny valley, where it's almost close enough but obviously doesn't have the "flow" that Carlin would have ensured. George was an absolute master of language and how to employ it, and his routines were like gallery pieces painstakingly and exactingly carved from ebony and bone, then posed and exhibited with excruciating care. This was more like a white-and-black plastic 3D printed version in a tourist gift shop's window. Close enough to make you double-take, but obvious when inspected.

--Patrick

I disagree with this assessment. Live comedy will always be something people go to see. HBO specials are something different, but even that would require a human to write the material. At least for now AI is absolutely unable to write meaningful social commentary that's original. Oh, they will absolutely steal material from others...

Jan 12, 2024#184

Ravenpoe

Dave said:
Oh, they will absolutely steal material from others...

Dear God, they've made Carlos Mencia

Jan 12, 2024#185

MindDetective

Ravenpoe said:
Dear God, they've made Carlos Mencia

To honor him?

Jan 13, 2024#186

figmentPez

Amazon has an AI problem

Jan 13, 2024#187

PatrThom

Dave said:
I disagree with this assessment.

As do I. For the record, I do not believe AI will mean the death of stand-up. I was merely describing which specific segment I thought you would find most relevant to your interests.

--Patrick

Jan 13, 2024#188

figmentPez

Microsoft wants to add an AI button to keyboards

They haven't yet decided if the "Copilot" key will replace the Windows key.

Jan 14, 2024#189

Dave

The jobs that AI could replace and nobody would care would be Board Members and CEOs. I mean, if you can be a Board Member on more than one board, it's not a job, it's a side gig. You can be replaced.

Jan 20, 2024#190

GasBandit

Jan 20, 2024#191

figmentPez

Chicago University has released the 1.0 version of Nightshade, software meant to "poison" artwork such that it will damage the database of any AI that attempt to use it.

Jan 20, 2024#192

PatrThom

Also a tool called "Glaze" to obfuscate your artistic style.
For people using both, 'shade it first, then Glaze it. It's more important to Glaze than 'shade.

Additional stuff in the news:

Game developer survey: 50% work at a studio already using generative AI tools

But 84% of devs are at least somewhat concerned about ethical use of those tools.

arstechnica.com

--Patrick

Jan 20, 2024#193

bhamv3

GasBandit said:

That's not even a good poem!

Jan 25, 2024#194

Frank

The slow death of art marches on.

https://edition.cnn.com/2024/01/19/style/rie-kudan-akutagawa-prize-chatgpt/index.html

Jan 26, 2024#195

Dave

George Carlin Estate Files Lawsuit Against Group Behind AI-Generated Stand-Up Special: ‘A Casual Theft of a Great American Artist’s Work’

George Carlin's estate has filed a lawsuit against the creators behind an AI-generated comedy special featuring a recreation of the comedian's voice.

variety.com

Jan 26, 2024#196

PatrThom

Did an AI write that hour-long “George Carlin” special? I’m not convinced.

"Everyone is ready to believe that AI can do things, even if it can't."

arstechnica.com

--Patrick

Jan 26, 2024#197

Dave

AI did not write that. AI changed it into Carlin's voice. AI's start good and tend to end up sounding like Trump.

Waitaminute....!

Jan 28, 2024#198

Dave

I may not be the funniest guy around, but I know comedy.

Following lawsuit, rep admits “AI” George Carlin was human-written

Creators still face "name and likeness" complaints; lawyer says suit will continue.

arstechnica.com

Jan 28, 2024#199

PatrThom

I could not believe the text was created by A.I.
The voice, sure. But not the routine itself.

--Patrick

Feb 5, 2024#200

GasBandit

Feb 13, 2024#201

GasBandit

Feb 13, 2024#202

PatrThom

Omitted red button text: "I didn't realize how traumatic reality is until I 100% checked out."

--Patrick

Feb 14, 2024#203

Bubble181

PatrThom said:
Omitted red button text: "I didn't realize how traumatic reality is until I 100% checked out."

--Patrick

And mouse over, "the really sad part is when they start deepfaking themselves for affirmation"

Feb 14, 2024#204

PatrThom

Thanks, I'll remember to include that next time someone else forgets.

--Patrick

Feb 16, 2024#205

PatrThom

Scientists aghast at bizarre AI rat with huge genitals in peer-reviewed article

It's unclear how such egregiously bad images made it through peer-review.

arstechnica.com

You need to click through to this if only for the A.I.llustration of the rat with the OUTRAGEOUS penis. It's practically Minotaur-sized. And it's labeled "dck."

--Patrick

Feb 17, 2024#206

Bubble181

Well, I mean, that kind of deliberate oversizing isn't totally abnormal for that sort of diagram. In biology textbooks you'd find dozens like that with blown up chest cavities or stomachs or whatever to be able to more clearly label and identify specific bits.
It's the totally random gibberish labels that do it for me... That means nobody looked at/corrected/checked them at all

Feb 18, 2024#207

figmentPez

I genuinely don't know if this is satire, or just clueless AI supporters being oblivious:

AI proponent complaining about someone stealing their prompt.png

Source tweet

This appears to be an A.I. proponent complaining about how someone generated a new image from a copied prompt and got more attention on the internet with it. They say that the image they prompted "is unique, there is no other like it. And I created it." but they're also livid that someone else "created" something very much like it.

This could be very on-the-nose satire, but I'm not sure how anyone would be able to tell.

Feb 18, 2024#208

GasBandit

"You're trying to kidnap what I have rightfully stolen."

Feb 18, 2024#209

figmentPez

This is a fascinating perspective on A.I. that I had not considered before:

Here's the Threadreader version.

The TL;DR is that it's from a teacher who found out students have been using ChatGPT to try to sound "more respectable" in the teacher's words and "more white" in the words of one of the students.

"The student also told me that in therapy, their therapist had been misunderstanding them, blaming them, and denying that these misunderstandings were because of a language barrier. They felt that they were so bad at communicating, because of their language, and their culture, and their head injury, that they would never be a good scholar. They thought they had to use chatGPT to make them sound like an American, or they would never get a job."

Students who were doing the work, and understood the subject, were then passing their writing through ChatGPT in order to make it seem more polished.

"They also told me that when they used chatGPT to help them write emails, they got more responses, which helped them with research recruitment. I've heard this from other students too. That faculty only respond to their emails when they use chatGPT."

EDIT: Forgot to mention that while ChatGPT is effecitve in making emails sound more professional, it backfires on the students when used for polishing research papers. It works well for pleasantries, but turns researched arguments into wordy gibberish.

Feb 18, 2024#210

PatrThom

GOODY-2 | The world's most responsible AI model

Introducing a new AI model with next-gen ethical alignment. Chat now.

www.goody2.ai

--Patrick

Feb 26, 2024#211

GasBandit

Feb 26, 2024#212

GasBandit

Feb 28, 2024#213

Gryfter

Have you all heard about the Willy Wonka Experience?

Spoiler it's not good.

Cops Called to 'Willy Wonka Experience' as Crying Children Realize AI Ads Were Lies

Willy Wonka Experience’s AI-fueled promises were pure imagination.

gizmodo.com

Actor From Willy Wonka Fiasco Says AI-Generated Script Was 'Terrifying for the Kids'

“I paid for Willy Wonka and got Billy Bonkers,” wrote one angry customer.

gizmodo.com

Feb 28, 2024#214

Dave

That's unintentionally hilarious. The voices of the kids when the "Unknown" came out..."Noooooooo....!"

And the Debbie Downer looking meth Oompa Loompa. Just comedy all around!

Feb 28, 2024#215

drifter

Did the same people who did Dashcon set this one up?

Mar 1, 2024#216

figmentPez

DocuSign has admitted that it uses customer data to train AI

https://bsky.app/profile/clancyny.bsky.social/post/3kml3nwvatd2f

DocuSign is using customer data to train AI.jpg

Huge amounts of legal documents being mashed together in a database being used by software for unknown purposes, without the consent of users. What could possibly go wrong?

Mar 1, 2024#217

Bubble181

figmentPez said:
What could possibly go wrong?

"Do you want the short list, or the long?"

Mar 1, 2024#218

PatrThom

Corporation don't care. Corporation has access to a giant corpus of text. Corporation is approached by A.I. corporation who says, "We will give you a tidy sum of money if you just let us read what you have there." Corporation just thinks, "Ooo, free money!"

--Patrick

Mar 5, 2024#219

evilmike

Dave said:
And the Debbie Downer looking meth Oompa Loompa. Just comedy all around!

The gift that keeps on giving:

Mar 5, 2024#220

GasBandit

Mar 6, 2024#221

Frank

My Robotic Doppelganger Is The Grim Face of Journalism's Future!

As journalism endures catastrophic financial collapse, I watch software seize my long-lost archives, erase my name—or attribute my work to a fake person. PLUS: Evan and I sign on Friday!

foreverwars.ghost.io

Mar 6, 2024#222

evilmike

Mar 16, 2024#223

PatrThom

AI-generated food images look tastier than real ones | University of Oxford

Researchers have announced an intriguing discovery – consumers generally prefer AI-generated images of food over real food images, especially when they are unaware of their true nature. The new findings have been published in Food Quality and Preference.

www.ox.ac.uk

Food Stylists: "They took our jerbs!"

--Patrick

Mar 16, 2024#224

figmentPez

I wanted to know how many square feet of pizza you get with a 42" pie, but I'm having a bad health day and wasn't about to trust my math, so I decided to try Google Gemini (formerly Bard) to see if it could get some fairly basic math right.

Spoiler, it could not.

$Google Gemini cannot do math.png$

Gemini even gave a pop-up claiming that it searched Google to verify that the results were correct.

Elsewhere, Wolfram Alpha gave the correct answer:

Wolfram Alpha gets the correct answer.png

9.6 versus 1,385.44

That's a pretty big difference.

Even telling Gemini that it got the wrong answer only causes it to redo the math in a different wrong way. Somehow it's getting this:

$Google Gemini cannot do math again.png$

3.14159 * 21^2 = 1,385.44119. I have no idea how it inflated that number by 2%, and that was even after telling it that 10.1 square feet was a wrong answer.

I'm glad I only put this into Gemini for a laugh, because those answers are just horrible.

Mar 16, 2024#225

PatrThom

Google doesn't understand PEMDAS.
This explains why it's so hard to build a decent search these days.

--Patrick

Mar 26, 2024#226

Tinwhistler

Just a short time ago, Rolling Stone interviewed some guys who said that AI music generation from generalized text prompts was "years away".
Spoiler alert: It wasn't.

I read an article about new developments, found a website typed in a general prompt, generated a song, threw that into Clipchamp with some pics and vids from my hard drive, and made this in like 10 minutes.

I'm not quite sure how I feel about it yet.

Mar 26, 2024#227

Dave

Only thing the song was missing was a penny whistle solo.

Mar 26, 2024#228

Tinwhistler

Dave said:
Only thing the song was missing was a penny whistle solo.

I agree...I've listened to it a bit to see if I can get it under my skin, which is really the only way I've ever been able to play something by ear without sheet music

Mar 27, 2024#229

GasBandit

Mar 27, 2024#230

GasBandit

Not sure about the album art, but hey, I've heard worse..

Accelerate by @rarebreakdance917 | Suno

fast-paced energetic eurobeat song. Listen and make your own with Suno.

app.suno.ai

Mar 27, 2024#231

Tinwhistler

GasBandit said:
Not sure about the album art, but hey, I've heard worse..

Accelerate by @rarebreakdance917 | Suno

fast-paced energetic eurobeat song. Listen and make your own with Suno.

app.suno.ai

I tried to make one that was vaguely pokemon theme song feeling for my dragon biz. It's very....autotuned. But I'm super impressed with how polished it is, even with the occasional ai-generated weirdness (the word "shapes" for instance).

Mar 30, 2024#232

PatrThom

@IronBrig4 @Tress

Anyone currently looking for a job, make sure you stay until the very end.

--Patrick

Apr 4, 2024#233

figmentPez

The Washington State Lottery put up a website that let people use AI to "test drive a win", making images of themselves if they won the lotto. A woman uploaded her picture and told it to make a pic of her dream vacation to swim with sharks. It came back with a pic of a topless woman, with her face, kneeling on a bed.

After AI-generated porn report, Washington Lottery pulls down interactive web app

I'm not at all surprised that people running the lottery were also the type to try to grift people using AI.

Apr 5, 2024#234

figmentPez

AI bros are trying to portray opposition as "AI phobia" and are trying to use #BreakThePencil as a slogan, saying "pencil is a slur", and claiming that AI "artists" are oppressed.

Or maybe it's just a troll meme trying to cause outrage. How could anyone possibly know? AI bros are always on the border of self-parody, and a lot of exploitative shitheads are appropriating the language of the oppressed lately.

Apr 5, 2024#235

Frank

figmentPez said:
AI bros are trying to portray opposition as "AI phobia" and are trying to use #BreakThePencil as a slogan, saying "pencil is a slur", and claiming that AI "artists" are oppressed.

Or maybe it's just a troll meme trying to cause outrage. How could anyone possibly know? AI bros are always on the border of self-parody, and a lot of exploitative shitheads are appropriating the language of the oppressed lately.

Gamergate self victimization playbook. Gamers are truly the most oppressed minority.

Apr 5, 2024#236

figmentPez

Fuck Pizza Hut "This ad was created with AI to save you money."

Pizza Hut is using AI generated ads and bragging about it.jpg

Apr 5, 2024#237

PatrThom

figmentPez said:
Fuck Pizza Hut "This ad was created with AI to save you money."

--Patrick

Apr 14, 2024#238

Bubble181

Apr 14, 2024#239

GasBandit

TLDR: Roko's Basilisk.

Apr 14, 2024#240

PatrThom

Bubble181 said:
View attachment 47837

Tooltip/mouseover text: "GREAT JOB AI, STAY WITH ART YOU'RE NAILING IT"
Red button text: "First rule of AI: no toothbrush mustaches."

--Patrick

Apr 19, 2024#241

PatrThom

Gamers are renting out GPU power for AI porn generation in return for Fortnite skins

According to a report from 404 Media, Salad has found a clever way to massively scale up AI content generation, including images, videos, and text, by outsourcing...

www.techspot.com

Aw, man. Nobody wants Robux any more.

--Patrick

Apr 19, 2024#242

figmentPez

No mention of electricity costs there. So I'm guessing they're either counting on gamer's parents to foot the bill, unknowingly, or they're hoping that their target group of gamers will have no idea they'll still be paying for this, one way or another.

Monday at 4:34 PM#243

GasBandit

Monday at 5:14 PM#244

Bubble181

GasBandit said:

Yep, nearly the same level of understanding and intelligence as some of my foreign colleagues.

Thursday at 12:43 PM#245

GasBandit

Forums