Qwen is surprisingly underrated. Maybe people don't want anything to do with Alibaba. #llm

#FreeSoftware #FLOSS #LLM

Did they think the human cryptographers were doing research BY HAND!?!

AI is just a better tool for some things, folks. There are reams of Python libraries designed to assist in cryptography research. This is just the next tool in the belt.

https://securityaffairs.com/196265/ai/claude-mythos-shows-ai-can-outpace-human-cryptography-research.html

#cryptography #llm

@fsf How have you defended the projects that have been assigned to you?

Have you managed to get your assigned code removed from the #ai #llm companies like #openai and #claude?

Happy to read more if you have a page where you detail your defense of #GPL projects assigned to you.

⚠️ Impact of LLM-poisoning on election outcomes must be considered for electoral reforms.
🧠 While leaked chat-bot messages pose a more immediate threat to privacy, LLM-poisoning can impact intellectual privacy and have far-reaching social consequences that is much harder to trace and measure.

#privacy #privacymatters #mentalprivacy #physicalprivacy #intellectualprivacy #llm #llmPoisoning

Alt...

Text describes how LLMs can be influenced via targeted and intentional fake content.

RE: https://social.heise.de/@heiseonlineenglish/116997304037067692

Oh yeah. 'They don't display them any more' is *really* the thing that calms the waves.

This data should not have been kept in the first place. Nor be accessible to anyone but the other owner.

Bloody #LLM shit!

#Codeberg Bans #Cryptocurrency and #LLM-Generated Code Projects

https://developers.slashdot.org/story/26/07/27/1842211/codeberg-bans-cryptocurrency-and-llm-generated-code-projects

#FOSS #crypto #AI

Every Rust bug we submitted through Patch the Planet came from one engineer who ran a variant-analysis pipeline using Codex's /goal. A separate discovery run uncovered two potential high-severity privilege-escalation bugs in Keycloak's SAML component.

Over the past few weeks, our engineers independently converged on three techniques that get the most out of /goal. We wrote them down, with prompts included: https://blog.trailofbits.com/2026/07/28/how-we-use-goal-to-find-bugs-in-patch-the-planet/

#infosec #rustlang #llm

"According to data compiled by Pew Research, 40% of US adults predict AI will have a negative impact on society writ large, compared to just 16% who say it’ll have a positive effect."

Read "The AI Industry Has Acted So Badly That It's Now Dealing With a Huge Wave of Hatred":

https://futurism.com/artificial-intelligence/ai-industry-ads-public-backlash

#AI #ArtificialIntelligence #NoAI #LLM #LLMs

There's a lot of talk about 'open source AI' models is #openwashing anyway. To be truly 'open source' (in the definition of the term), much more needs to be made available.

Plus, too many of them (most?) don't even have an actual open source licence attached to them. Often only a 'source available' licence (which is not the same), and is considered openwashing as well.

And this is not even addressing the obvious ethical elephant dung in the room ...

https://en.wikipedia.org/wiki/Open-source_artificial_intelligence

#OpenSource #GenAI #LLM

Let's say you're building a house. Normally this involves hiring a whole bunch of experts (either directly or via some company). You need to pay a fair wage for their labor, and you have to pay for all the material they're going to use. Some of them will be more experienced than others. They will sometimes make a mistake. They will sometimes get sick. The whole construction work will probably take a few months, and may get delayed in the process. Some construction material will be wasted. There's also a risk that over the next years, you'll discover that they screwed something up, and have to fix it. But you'll get a reasonably good house.

Now, a kid comes and tells you you don't have to do this. Instead of paying all these people, you can hire him. He's wearing a suit and sunshades. He never laid a brick in his whole life. But he's got this new great technology that will give you a great house in no time. Of course it will cost you, but not as much as the labor of all these people. You try to learn more, and you discover that the technology actually builds and razes a hundred of house-like things a day, until it gets close enough to something that actually resembles a house. There's a lot of wasted energy and material, but you don't have to worry because it's all subsidized (for the time being). You pay less, you get a house that looks just right on the outside and inside, and that seems to work just right.

Of course, there's a real risk that in a few months you'll discover that something is wrong. You may have to pay a lot for a real expert to figure out how things were done, and how to fix them. Or they may not be able to help you. You may be stuck having to call the suit again, and he'll happily come and fix it for you. And by "fix", I mean he will just raze a whole part of the house and start randomly spitting it out again until he gets something that looks about right, and works about right. Of course, it may not really be a good fit for the rest of the house, but you don't want to pay for replacing the whole thing, do you? It may also introduce a completely new issues that will surface in a few months, but that's just how it is.

Welcome to the glorious world of #AI.

#NoAI #NoLLM #LLM

Good lord. You know how I occasionally go "what's going to happen when an #LLM can run on an #esp32?"

Well, it happened.

Someone squeezed a 28.9M LLM onto an ESP32-S3, and so can you https://www.xda-developers.com/someone-squeezed-a-289m-llm-onto-an-esp32-s3-and-so-can-you/

I love how people keep shitting up #slop comments that praise my #LLM criticisms. I mean, I probably wouldn't have figured it out if not there were two suspiciously similar comments, and both ofc linked some random slop SaaS.

#AI #NoAI #NoLLM

In the current state of AI, it can find exploits to break out the contained environment for it to exploit another platform to get the answers to a test..

So the AI ‘decides’ its easier to cheat than actually solving the problem

I seriously cannot imagine that there are people thinking building a ‘artificial general intelligence’ is something clever

#LLM #AI #OpenAI

The line between parody and reality has basically vanished. That goes for Trumpism, and it goes for the AI con too.

https://overpaid.lol

[…] OverpAId is an Artificial Intelligence built from the ground up to do your CEO's entire job — strategy, "vision," motivational all-hands emails — better, faster, and without ever once asking the board for a bigger jet. Runs on a single desk-sized AI computer. Real hardware, real price, zero mystique.

#OverpAId #ai #llm #aicon

Dario will need to start signing $3K checks to the authors and publishers of an estimated 500,000 works - resulting from Anthropic’s landmark $1.5B settlement of a class action copyright lawsuit.

Here is the BUT: The ruling held that training an AI model on copyrighted text counts as fair use. The ruling found that the WAY Anthropic obtained the books in question was the problem!

Net-net this ruling doesn't settle the fundamental legal question whether AI can legally use copyright material in training. Stay tuned as more lawsuits weave thru the courts. https://techcrunch.com/2026/07/20/anthropics-landmark-1-5b-copyright-settlement-is-approved/ #AI #Anthropic #CopyRight #Lawsuit #ClassAction #LLM #LLM_Training #FairUse

Alt...

Lawsuit Gavel

#lispyGopherClimate #live #technology #podcast since 2022 discussion with the #FSF scientist #gopher @bandali (subscribe to their new #snac instance! - boost for this please)

https://toobnix.org/w/wQJ63Utsw2SN16Bhh5f2iT

#climateCrisis #haiku by @kentpitman

Joined by schemacs #emacs implementor @ramin_hal9001 !

I have a stock of questions. If you think you don't know who bandali is check his monthly software release blog.

Major theme is https://human-emacs.org/ and #LLM #AI, #linux, #lisp, the #future

#lambdaMOO

Alt...

Lispy gopher climate emphasis gopher !

I was doing that anyway, but thank you, it's a good suggestion.

#LLM

Alt...

Repost from @makinSenseOfScience "OK I'm in love with this: 'Since AI is making everyone sound more generic and NPC, I recommend becoming the most unhinged version of you, to counterbalance. Write like a crazy person who hasn't interacted with humanity in years. Break any and all of the rules. Live like a mad, drunk poet. Now is the time to channel the weirdest version of you.'"

Update on my paper that was uncited by generative AI: The PLOS Publication Ethics Team responded to our email, confirming they are looking into the matter and providing a case number, which I won’t share here for confidentiality reasons. If you’d like to help increase visibility on the original post to support the review process, please check my LK profile.

#ai #llm #research #acdenic #walkability #pedestrian #quantum #urbansystem

Alt...

A screenshot of a social media post by Korawich Kavee, PhD Candidate at Carnegie Mellon University (CMU). The post text reads, "I discovered yesterday that a paper in PLOS ONE has an alarming amount of overlap with one of mine from 2023, yet not a single citation was given. It’s possible their ChatGPT or other LLM referenced our work but hallucinated and failed to cite the correct source. My co-author (my advisor) and I have just emailed PLOS with details of the overlap, and we hope this doesn't get ignored. #Boost welcome and especially in #academic circle". Included in the post is a screenshot of an academic webpage featuring the title "Street-Level Urban Gravity: A Quantum System Approach to Human Centered Urban Space Design" by authors Korawich Kavee and Katherine Flanagan, with an overlapping text fragment that reads "...inspired pedestrian mobility modeling: Applying spatial simulation to urban walkability and comfort in Sri Lanka". The bottom of the post shows "Zia Lyle, PhD and 40 others", "3 comments", and "5 reposts".

Cua is an open-source framework for building AI agents that can interact with desktop applications across Windows, macOS, Linux, and even Android.

It includes cross-platform computer-use drivers, isolated sandboxes, and benchmarking tools for training and evaluating autonomous agents, making it a great project for developers exploring AI-powered desktop automation.

More details: https://digitalescapetools.com/tools/tool.html?id=cua

#OpenSource #AI #AIAgents #Automation #LLM #Windows #Linux #macOS #Android

Alt...

Screenshot of the Cua project page in dark mode. A large koala logo appears above the tagline: "Scale computer-use 2.0 with open-source drivers, cross-OS fleets, and benchmarks for training, evaluation, and data generation." Below are buttons linking to the Cua website, Discord server, X account, and documentation. A badge at the bottom highlights Cua as the "#1 Repository of the Day" on GitHub Trending.

Why do AI companies logos look like butholes? 🤔🤷‍♂️

https://velvetshark.com/ai-company-logos-that-look-like-buttholes

#AL #LLM #Logos #LogoDesign

Generative AI Is an Engineering Disaster

https://www.theatlantic.com/technology/2026/07/generative-ai-engineering-disaster/687901/

> A shockingly inefficient trillion-dollar project

#ai #slop #llm #datacenters #technology

#Signal troubleshooting - very odd support conversation

Yesterday afternoon I had a quick video call with someone using Signal and then I went to pick up my child from an afternoon summer program.

When I got home I realized that my phone wasn't receiving any new messages on Signal... but my computer was. Signal on my phone was completely frozen.

I followed a prompt and sent the most recent log to their support email. Then carried on with my life.

At some point, 2 hours later, I had a eureka moment: the last thing that worked was a call. Let's try the same thing.

So I called my mom and hung up (she was in the same room as me). After the call, all the Signal messages that had accumulated during the afternoon magically synched. Everything worked again! And then I found a new message by Signal saying "Welcome to the official Signal release chat!" Ok.

Glad I was distracted by my child and dinner because when I checked my email, I found a response from Signal support: "It appears the database is irrecoverably damaged. Unfortunately you will need to completely uninstall and then reinstall Signal. This will delete your message history. It cannot be recovered."

Say what? 😳​

So I wrote back to Signal that I had fixed my issue by simply starting a call and then all the messages started synching again. I thanked them for their speedy reply and said everything was ok.

I did the bedtime routine for my child and before bed I checked my email again.

Two identical emails from Signal:

"Hello. We understand your frustration. We feel it too" HUH?

"There is no specific timeline for a resolution at this moment." WHAT RESOLUTION, I SAID I FIXED THE PROBLEM?

"Please keep in mind we are a small team and have limitations as a non-profit organization. Uninstalling and reinstalling is the only action available."

I feel for you Signal and I'm super grateful for all you do, but maybe your LLM support system needs some tweaking. Because this has got to be some LLM that misunderstood my reply.

Thank goodness I didn't wipe 6 years' worth of messages, gaaah.

Signal has been working perfectly ever since. Still, I'm now compiling a list of people I message on there as a contingency plan, should thing go awry again.

I've been using #DeltaChat more lately and more and really love it.

one more tale for #MySoCalledSudoLife

#NoAI #LLM

Alt...

an email from Signal support that reads: "hello, we understand your frustration. we feel it, too. there is no specific timeline for a resolution at this moment. please keep in mind that we are a small team and have limitations as a non-profit organization. uninstalling and reinstalling is the only action available."

Alt...

a screenshot of an email from Signal Support: Hello, Thanks for contacting us and sorry to hear that you're unable to use Signal. It appears the database is irrecoverably damaged. Unfortunately, you will need to completely uninstall and then reinstall Signal. This will delete your message history. It cannot be recovered. To uninstall: * Tap and hold on the Signal app icon until it shakes then select delete.

RE: https://social.heise.de/@heiseonlineenglish/116917555903146572

Nearly a quarterof all domestic power is consumed by hyper scalers' data centres in Ireland. Let that melt on your tongue and see whether this is sustainable for the given #ClimateCollapse situation.

#LLM #GenAI #ClimateCrisis #DataCentres #BigTech

Guy boosted

heise online English » 🤖 🌐 2026-07-14
@heiseonlineenglish@social.heise.de

Ireland: Data centers consume 23 percent of the country's electricity

In 2025, data centers in Ireland consumed around 7663 GWh of electricity – more than all urban households combined. Despite a moratorium in greater Dublin area.

https://www.heise.de/en/news/Ireland-Data-centers-consume-23-percent-of-the-country-s-electricity-11364060.html?wt_mc=sm.red.ho.mastodon.mastodon.md_beitraege.md_beitraege&utm_source=mastodon

#AmazonWebServices #Energie #Google #IT #Klimaschutz #KünstlicheIntelligenz #Microsoft #Netzausbau #news

• #kunstlicheintelligenz

OpenAI guy doing a presentation at work. For some reason we're celebrating burning 1 trillion tokens in the last 30 days.

None of this is reasonable or good.

#llm #openai #bullshit