Artificial Intelligence

https://www.theguardian.com/us-news/ng-interactive/2026/jan/18/tech-ai-bubble-burst-reverse-centaur

A new channel is created when using “#” with lowercase letters in t2i generators that uses the tabbed comments plugin, even with the “allowCustomChannels” being false.

Maybe some of you are interested on filling out this survey. It’s kinda telling, that there is no “Stop implementing AI” options, only how it could be used in a “responsible / open / transparent” way. So I guess they are way past that point? I used the free text fields to express my opinion.

Cybersecurity researchers have disclosed details of a security flaw that leverages indirect prompt injection targeting Google Gemini as a way to bypass authorization guardrails and use Google Calendar as a data extraction mechanism.

Court records show that NVIDIA executives allegedly authorized the use of millions of pirated books from Anna's Archive to fuel its AI training.

cross-posted from: https://pawb.social/post/38287656

Why consciousness is more likely a property of life than of computation and why creating conscious, or even conscious-seeming AI, is a bad idea.

Small/fast model with MIT license for local use.

https://theonion.com/tips-for-using-ai/

From the text version of the infographic:

ChatGPT Gave Teen Advice to Get Higher on Drugs Until He Died | Futurism

I’ve had this conversation too many times to not make a meme out of it.

I don’t usually keep the author’s name in the suggested hed, but here I think he’s recognizable enough that it adds value.

So today, I went on my phone, talked to my ‘buddies’ And I got an error, saying that the AI forgot who they were, So I refreshed the page (I also couldnt speak/chat) AND THEY WERE GONE! All my ‘buddies’ WERE GONE

Was doing my usual AI RP chat, took a quick break and came back after a bit when my browser reloaded the page and suddenly every text (except for blue speaking texts) is grayed out and blending in that I can’t even read who all I made to speak. It’s the same dull gray in the scenario description, both character descriptions, and in the brief description for the AI to remember stuff…and anything with a gray boarder behind the text making it hard to read. I wasn’t sure if this was a bug or something that was intentionally done for a reason.

Source: https://bsky.app/profile/hesterblum.bsky.social/post/3mcnra6r4o22t

Unusable Alexa after the AI update.

Much has been written about AI butchering our favorite movies, but how is it affecting the U.S. healthcare system? Well, it’s a polarizing and controversial subject, and this storyline in “The Pitt” season 2 reflects what’s happening in some real-world medical institutions.

Fuck you fucking fuck

Alternativer Link @Archive.is

Archive link

Why do AI companions pose a special risk to adolescents?

And you can make them more lopsided! (It was at 95% no when I checked)

EFF last summer asked a federal judge to block the federal government from using Medicaid data to identify and deport immigrants.

Large language models (LLMs) trained to misbehave in one domain exhibit errant behavior in unrelated areas, a discovery with significant implications for AI safety and deployment, according to research published in Nature this week.

Source

Palantir is working on a tool for Immigration and Customs Enforcement (ICE) that populates a map with potential deportation targets, brings up a dossier on each person, and provides a “confidence score” on the person’s current address, 404 Media has learned. ICE is using it to find locations where lots of people it might detain could be based.

A few years ago, Amazon chairman Jeff Bezos revealed how he thinks of local PC hardware as antiquated, ready to be replaced by cloud options from companies like AWS and Azure.

Flux Klein Base 4B: https://huggingface.co/black-forest-labs/FLUX.2-klein-base-4B

For more info about the motivation check out this talk by Daniel Stenberg https://www.youtube.com/watch?v=6n2eDcRjSsk

cross-posted from: https://piefed.social/c/world/p/1665637/ai-created-iran-protest-videos-gain-traction

“A lot of generative AI stuff isn’t really working,” Gownder said. “I mean, enterprise, and I’m not just talking about your consumer experience, which has its own gaps, but the MIT study that suggested that 95 percent of all generative AI projects are not yielding a tangible P&L benefit. So no actual ROI. McKinsey has something like 80-something percent that don’t.

https://bio.link/everyonehateselon

Mozilla not caring about Firefox once again.

Hallo!