AI don't trust techbros

[abates]

Topically relevant to this site!

I was watching a YouTube video about Challenge of the GoBots. A short way in, it mentioned that the show cast Peter Cullen and Frank Welker in parts, and after their mug shots, flashed up this image to demonstrate who they'd voiced in Transformers:

The image on the left, you probably don't recognise. I had to image search to work out that it is an AI image someone generated two years ago and uploaded to a website with the label "Optimus Prime 80s Cartoon". A round of sarcastic applause anyone?

 

[NovaSaber]

Cause and effect:

 

[CoffeeHorse]

There are still some good people left at Microsoft, and I feel sorry for them.

 

[Cybersnark]

Cops Forced to Explain Why AI Generated Police Report Claimed Officer Transformed Into Frog

AI-powered software Draft One automatically generates police reports from body camera footage. It thought an officer turned into a frog.

futurism.com

 

[CoffeeHorse]

Despite the drawbacks, Keel told the outlet that the tool is saving him “six to eight hours weekly now.”

“I’m not the most tech-savvy person, so it’s very user-friendly,” he added.

HIRE YOUNG PEOPLE FOR SOMETHING FOR GOD'S HUGGING SAKE!

 

[Pocket]

You're behind the times. It's the young people who don't know anything about technology now.

And I highly doubt the police force is mostly made up of boomers.

 

[Dekafox]

https://archive.ph/pINs8

 

[NovaSaber]

Musician's show gets canceled after Google AI incorrectly calls him a sex offender

AI search results on Google's terrible overview feature wrongly labeled Canadian musician Ashley MacIsaac a sex offender, resulting in his concert’s cancellation.

popculturenews.com

According to MacIsaac, the AI-generated summary falsely claimed he had been convicted of several serious crimes, including sexual assault, internet luring, assaulting a woman, and attempting to assault a minor. The summary also incorrectly stated that he was listed on Canada’s national sex offender registry — allegations MacIsaac says are entirely untrue.

MacIsaac later learned the false claims stemmed from online articles involving a different individual in Atlantic Canada who shares the same last name. He said the AI system appeared to have conflated those reports with his own public profile.

“If a lawyer wants to take this on (for free) … I would stand up because I’m not the first and I’m sure I won’t be the last,” he said.

 

[abates]

A high-scoring Reddit post about how Uber Eats screws over its drivers turned out to be a hoax entirely created using AI:

Debunking the AI food delivery hoax that fooled Reddit

A “whistleblower” tried to corroborate his viral post with AI-generated evidence. This is how I caught him. PLUS: Grok's image-generation crisis, and the rapture over Claude Opus 4.5

www.platformer.news

 

[Pocket]

OK but how much you want to bet Uber and DoorDash both took notes on that post and are figuring out how to implement all those things as soon as possible? I mean, if everyone already believes you're that evil, might as well deserve it, right?

 

[CoffeeHorse]

A lot of the Redditors who got fooled by this story are probably AI too. Reddit is a mess.

 

[wonko the sane?]

It's not really "fooling them" when they exist solely to propagate the story. I'm almost certain clankers can tell when another clanker posts.

 

[Dekafox]

ChatGPT Health lets you connect medical records to an AI that makes things up

New feature will allow users to link medical and wellness records to AI chatbot.

arstechnica.com

The modern WebMD, with new added "medical data breach via prompt injection" bonus feature!

 

[CoffeeHorse]

Guess who's getting a job in the Pentagon!

WASHINGTON (AP) — Defense Secretary Pete Hegseth said Monday that Elon Musk’s artificial intelligence chatbot Grok will join Google’s generative AI engine in operating inside the Pentagon network, as part of a broader push to feed as much of the military’s data as possible into the developing technology.

“Very soon we will have the world’s leading AI models on every unclassified and classified network throughout our department,” Hegseth said in a speech at Musk’s space flight company, SpaceX, in South Texas.

[...]

Hegseth said Grok will go live inside the Defense Department later this month and announced that he would “make all appropriate data” from the military’s IT systems available for “AI exploitation.” He also said data from intelligence databases would be fed into AI systems.

Pentagon is embracing Musk's Grok AI chatbot as it draws global outcry

Defense Secretary Pete Hegseth says Elon Musk’s artificial intelligence chatbot Grok will join Google’s AI engine inside the Pentagon network.

apnews.com

What could go wrong?

 

[abates]

It will be much quicker to ask Grok on Twitter about Pentagon secrets than going to the trouble of buying the secret documents off Trump, I guess.

 

[KidTDragon]

And get some fake porn at the same time. No point in making two trips, right?

 

[wonko the sane?]

I mean, I figured it was there for the porn first. Make it easy for the regime.

 

[Pocket]

"Hey Grok, generate rule 34 of the Pentagon's current war plans."

 

[NovaSaber]

 

[Pocket]

The bananapants thing is that techbros managed to actually invent the Magic Button That Just Summons Finished Products Out Of Thin Air and they're having a hard time figuring out how to sell it to people.

Does kind of improve my faith in humanity at least.