Nobody cares about "AI" (Chatbot: I disagree.)

[Noodles]

I want to get medical advice from the robot nurse about anxiety induced by robot nurses.

 

[Kamilah Hauptmann]

A neat thing about text to image is how it highlights what is a commonly misspelled word.

Like when you ask for a prostrate figure and it gives you anal prolapses.

 

[Bartholomew Gallacher]

The CEO of Stability AI, Emad Mostaque, resigns as CEO of the company to pursue decentralised AI (which of course is bullshit). Stability AI is the company behind Stable Diffusion. He left after several key people resigned before him.

Stability AI CEO resigns to “pursue decentralized AI”

There was a lot of drama in the AI startup world this week.

www.theverge.com

 

[CronoCloud Creeggan]

Like when you ask for a prostrate figure and it gives you anal prolapses.

Facepalm.

 

[Argent Stonecutter]

LOL. This guy posted on Reddit about Claude making confident responses to being presented this MRI when actual doctors weren't willing to say anything was there, and commenter after commenter go "See. Claude is smarter than the doctor." If you go to the source post on Twitter the guy goes on to say that Claude was actually wrong, there really was no tumor there, it was just generating plausible responses because finding a tumor was more common in its training set than not finding one so that was a more plausible response. It wasn't actually doing any analysis at all.

And commenters on the thread are still talking about how it made an "educated guess" and how it was probably right and completely failing to understand what actually happened. I am so done.

 

[Free]

 

[Argent Stonecutter]

Spicy autocomplete has been fooling people since the '60s when Weizenbaum's own receptionist spent hours talking to Eliza.

 

[Free]

Spicy autocomplete has been fooling people since the '60s when Weizenbaum's own receptionist spent hours talking to Eliza.

OK, but then Eliza didn't ship you shit clothing and charge your Mastercard for the benefit.

 

[Cristiano]

Google's Trying Out AI to Help You Make Custom Travel Itineraries

If you want to check it out, you'll have to sign up for Google's Search Labs.

www.cnet.com

 

[Free]

 

[Free]

Oops! I guess it was sort of on topic, but that last post was meant to be this video:

 

[Free]

Welcome to the AI gadget era

The AI iPhone isn’t here yet — but is this the AI iPod era?

www.theverge.com

Remember the age of the internet appliance? I get a similar vibe from this.

 

[Veritable Quandry]

It was just a matter of time. And just like any other device that requires a specific service to function, i will avoid these.

 

[Noodles]

The CEO of Stability AI, Emad Mostaque, resigns as CEO of the company to pursue decentralised AI (which of course is bullshit). Stability AI is the company behind Stable Diffusion. He left after several key people resigned before him.

Decentralized AI? Like Blockchain?

Hey, let's take this energy eating monster, and infused it with our old energy eating monster!

 

[Noodles]

And commenters on the thread are still talking about how it made an "educated guess" and how it was probably right and completely failing to understand what actually happened. I am so done.

Yeah, the real danger and problem with AI. We already exist in a world where at least half the population believes lies that follow their desired worldview. Now we have robots that will tell us all the lies we can swallow, endlessly.

 

[Khamon]

Colossus only has to turn a few lights off and on and off and on.

 

[Dakota Tebaldi]

LOL. This guy posted on Reddit about Claude making confident responses to being presented this MRI when actual doctors weren't willing to say anything was there, and commenter after commenter go "See. Claude is smarter than the doctor." If you go to the source post on Twitter the guy goes on to say that Claude was actually wrong, there really was no tumor there, it was just generating plausible responses because finding a tumor was more common in its training set than not finding one so that was a more plausible response. It wasn't actually doing any analysis at all.

And commenters on the thread are still talking about how it made an "educated guess" and how it was probably right and completely failing to understand what actually happened. I am so done.

Reminds me of a similar kind of thing I'd read about, where an AI was being trained to diagnose tuberculosis from x-rays. Evidently it got quite decent at making correct diagnoses, but it turned out that since most of the training and test x-rays showing positive tuberculosis results had comes from poor-ish South and Central American countries (where tuberculosis is more common) with older and less-advanced equipment, the AI had "learned" that the visible hallmarks of an x-ray image from an older machine indicated probable tuberculosis.

 

[Argent Stonecutter]

Reminds me of a similar kind of thing I'd read about, where an AI was being trained to diagnose tuberculosis from x-rays. Evidently it got quite decent at making correct diagnoses, but it turned out that since most of the training and test x-rays showing positive tuberculosis results had comes from poor-ish South and Central American countries (where tuberculosis is more common) with older and less-advanced equipment, the AI had "learned" that the visible hallmarks of an x-ray image from an older machine indicated probable tuberculosis.

I remember that story. And that was a legitimate machine-learning scheme with positive and negative examples and a clear testable output... not a parody generator using the image as a prompt.

 

[Innula Zenovka]

“This is unparalleled, in my memory,” said one intelligence officer who used Lavender, adding that they had more faith in a “statistical mechanism” than a grieving soldier. “Everyone there, including me, lost people on October 7. The machine did it coldly. And that made it easier.”

Another Lavender user questioned whether humans’ role in the selection process was meaningful. “I would invest 20 seconds for each target at this stage, and do dozens of them every day. I had zero added-value as a human, apart from being a stamp of approval. It saved a lot of time.”

And that's the problem. While the notorious NKVD troikas of Stalin's pre-war purges, usually comprising the local NKVD chief, the local chief prosecutor and a representative of the local Communist Party, worked to targets of condemning so many thousand Kulaks/Ukrainians/Poles/whoever a week to death or the gulag for anti-Soviet activities, which generally left them around a minute to deliberate on each victim's fate, at least there was some human intervention and human responsibility.

While I quite see that AI is likely to be better at guessing who is a potential low-level Hamas operative than is a grieving soldier, it also takes away all human agency for these extra-judicial killings (for that is what they are) and leaves no one with the burden of responsibility.

 

[Free]

AI hype invades Taco Bell and Pizza Hut

Everything is suddenly "AI" in corporate food marketing, and we may have hit peak buzz.

arstechnica.com

Depending on who you ask about AI (and how you define it), the technology may or may not be useful, but one thing is for certain: AI hype is dominating corporate marketing these days—even in fast food. According to a report in The Wall Street Journal, corporate fast food giant Yum Brands is embracing an "AI-first mentality" across its restaurant chains, including Taco Bell, Pizza Hut, KFC, and Habit Burger Grill. The company's chief digital and technology officer, Joe Park, told the WSJ that AI will shape nearly every aspect of how these restaurants operate.

"Our vision of [quick-service restaurants] is that an AI-first mentality works every step of the way," Park said in an interview with the outlet. "If you think about the major journeys within a restaurant that can be AI-powered, we believe it’s endless."

I for one do not want to be serviced by our AI overlords.