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

[Free]

The future of having AI do tasks for you is here!

OpenAI's AI Agent Has to Be Monitored Nonstop to Catch Its Constant Screwups

OpenAI's new AI agent, Operator, isn't trustworthy or reliable enough to work without constant user supervision.

futurism.com

OpenAI has finally debuted "Operator," its very own AI agent, a type of autonomous model designed to do digital tasks on your behalf like shopping for groceries online.

But calling it an AI "toddler" might be more fitting. As a Bloomberg reporter describes her experience using OpenAI's new toy, the experimental tech needs a "lot of adult supervision," frequently screwing up and asking for help when it gets stuck.

It's also pretty sluggish, as echoed by other users, and slow on the uptake — as a still-developing brain might be.

"For several agonizing moments, I watched as OpenAI's artificially intelligent agent slowly navigated the internet like someone who's had the web described to them in great detail but never actually used it," wrote Bloomberg's Rachel Metz. "I had to monitor it the entire time."

I think I'd sooner expect a pet to learn how to walk itself.

 

[Noodles]

/me puts cigar to mouth "BRILLIANT!"

Add F*cking to Your Google Searches to Neutralize AI Summaries

Google will probably close this loophole soon enough.

gizmodo.com

modern problems

An image tagged modern problems

imgflip.com

 

[Bartholomew Gallacher]

Google decided to be even more evil, and silently dropped its rule to not develop AI powered weapons.

Google ending AI arms ban incredibly concerning, campaigners say

The tech giant previously ruled out applications that were "likely to cause harm".

www.bbc.com

 

[Argent Stonecutter]

It's secretly sabotage. Weapons using Google's so-called AI will actually reduce military effectiveness.

 

[Innula Zenovka]

https://existentialcomics.com/comic/588

 

[Free]

Meta's legal defense on how they trained their AI take a huge turn for the worse.

“Torrenting from a corporate laptop doesn’t feel right”: Meta emails unsealed

Meta's alleged torrenting and seeding of pirated books complicates copyright case.

arstechnica.com

Newly unsealed emails allegedly provide the "most damning evidence" yet against Meta in a copyright case raised by book authors alleging that Meta illegally trained its AI models on pirated books.

Last month, Meta admitted to torrenting a controversial large dataset known as LibGen, which includes tens of millions of pirated books. But details around the torrenting were murky until yesterday, when Meta's unredacted emails were made public for the first time. The new evidence showed that Meta torrented "at least 81.7 terabytes of data across multiple shadow libraries through the site Anna’s Archive, including at least 35.7 terabytes of data from Z-Library and LibGen," the authors' court filing said. And "Meta also previously torrented 80.6 terabytes of data from LibGen."

"The magnitude of Meta’s unlawful torrenting scheme is astonishing," the authors' filing alleged, insisting that "vastly smaller acts of data piracy—just .008 percent of the amount of copyrighted works Meta pirated—have resulted in Judges referring the conduct to the US Attorneys’ office for criminal investigation."

 

[Free]

An interesting, and free, online course about AI - not for the coders!

Carl T. Bergstrom (@carlbergstrom.com)

This is not a computer science course. It’s a humanities course about how to learn and work and thrive in an AI world. Neither instructor nor students need a technical background. Our instructor guide provides a choice of activities for each lesson that will easily fill an hour-long class...

bsky.app

Modern-Day Oracles or Bullshit Machines: Introduction

A free online humanities course about how to learn and work and thrive in an AI world.

thebullshitmachines.com

 

[Khamon]

Meta's legal defense on how they trained their AI take a huge turn for the worse.

Curious to see whether or not Meta is treated as badly as IA has been for providing limited and timed checkouts of digitized material.

 

[Bartholomew Gallacher]

There is hope, friends: Anthropic, the developer of the conversational AI assistant Claude, doesn’t want prospective new hires using AI assistants in their applications, regardless of whether they’re in marketing or engineering.

https://www.404media.co/anthropic-claude-job-application-ai-assistants/

 

[Free]

I think any researcher with 2 working neurons to rub together could have made this evaluation without much effort.

Study Finds That People Who Entrust Tasks to AI Are Losing Critical Thinking Skills

Trusting AI over the real thing seems, per a new study, to actually be atrophying folks' mental faculties.

futurism.com

Trusting artificial intelligence over the real thing seems, per a new study, to be atrophying folks' critical thinking skills.

As flagged by the folks at 404 Media, new research from Carnegie Mellon and Microsoft — yes, the same company that invested nearly $14 billion into OpenAI and is essentially subsidizing the ChatGPT maker — suggests that the more people use AI, to less critical thinking they do.

"Used improperly, technologies can and do result in the deterioration of cognitive faculties that ought to be preserved," the researchers wrote in the paper. "A key irony of automation is that by mechanising routine tasks and leaving exception-handling to the human user, you deprive the user of the routine opportunities to practice their judgement and strengthen their cognitive musculature, leaving them atrophied and unprepared when the exceptions do arise."

 

[Free]

Hoorah.

Thomson Reuters wins AI copyright ruling over training data

You gotta fight ... for your Reuters ... to party

www.theregister.com

Thomson Reuters has won a partial summary judgment in a copyright case against shuttered AI firm Ross Intelligence, a decision that disallows fair use as a defense for training models on proprietary data without permission.

"We are pleased that the court granted summary judgment in our favor and concluded that Westlaw’s editorial content created and maintained by our attorney editors, is protected by copyright and cannot be used without our consent," a spokesperson for Thomson Reuters told The Register today.

"The copying of our content was not 'fair use.'"

Ross Intelligence shut down years back, likely due to the lawsuit. But a win is a win.

 

[Free]

What do you call a thousand lawyers using AI to write their legal filings? The start of a sad, sad story, I'd say.

Attorney faces sanctions filing fake cases dreamed up by AI

Talk about court red-handed

www.theregister.com

Demonstrating yet again that uncritically trusting the output of generative AI is dangerous, attorneys involved in a product liability lawsuit have apologized to the presiding judge for submitting documents that cite non-existent legal cases.

The lawsuit began with a complaint filed in June, 2023, against Walmart and Jetson Electric Bikes over a fire allegedly caused by a hoverboard [PDF]. The blaze destroyed the plaintiffs' house and caused serious burns to family members, it is said.

Last week, Wyoming District Judge Kelly Rankin issued an order to show cause [PDF] that directs the plaintiffs' attorneys to explain why they should not be sanctioned for citing eight cases that do not exist in a January 22, 2025 filing.

To quote the attorneys acknowledging their error (that is, their error in relying on AI to not make shit up):

This serves as a cautionary tale for our firm and all firms, as we enter this new age of artificial intelligence.

But will anyone learn?

 

[Argent Stonecutter]

This is not a "new age of artificial intelligence". Idiots.

 

[Free]

This is not a "new age of artificial intelligence". Idiots.

The new age of acting artificially intelligent?

 

[Dakota Tebaldi]

What do you call a thousand lawyers using AI to write their legal filings? The start of a sad, sad story, I'd say.

Man....imagine being the family whose house burned down, who had what probably should have amounted to a slam-dunk case for recovering some damages, only to have it maybe irreparably ruined because the attorneys they hired turned out to be lazy bastiges who read that one article a year ago about ChatGPT "passing a bar exam" and thought they didn't have to do any work anymore.

 

[Free]

"True Crime" YouTube Channel Busted as Pure, Made-Up AI Slop

In a bizarre performance art project, a YouTuber profited from a "true crime" channel that was pure AI slop.

futurism.com

The creator of the now-defunct "True Crime Case Files" YouTube channel admitted to 404 Media that he intentionally dialed up the prurient interest on monetized videos with titles like "Coach Gives Cheerleader HIV after Secret Affair, Leading to Pregnancy."

According to Paul, the pseudonym given to the creator by 404 to protect his identity, the videos' scripts were half generated by OpenAI's ChatGPT and half written by him, and the visuals came from an AI image generator whose name he didn't disclose.

In what he says was an attempt to tip off viewers to the phony nature of the entire gambit, the creator would include strange details into his often-hypersexual storylines and name characters outlandishly.

So, the "Not True Crime Case Files" youtube channel.

 

[Kamilah Hauptmann]

 

[Khamon]

ha ha ha it thought you meant Hamburg, Alabama; well no, still wrong

 

[Free]

Ha ha ha. Someone is using Bing. Hilarious.

 

[Innula Zenovka]

AI cracks superbug problem in two days that took scientists years

The lead researcher has told the BBC he was so astounded he assumed his computer had been hacked.

www.bbc.co.uk