OpenAI Has Trained Its LLM To Confess To Bad Behavior
An anonymous reader quotes a report from MIT Technology Review: OpenAI is testing another new way to expose the complicated processes at work inside large language models. Researchers at the company can make an LLM produce what they call a confession, in which the model explains how it carried out a task and (most of the time) owns up to any bad behavior. Figuring out why … ⌘ Read more

Microsoft Faces New Complaint For Unlawfully Processing Data On Behalf of Israeli Military
Ancient Slashdot user Alain Williams shares a report from Al Jazeera: The Irish Council for Civil Liberties (ICCL) has announced it filed a complaint against Microsoft, accusing the global tech giant of unlawfully processing data on behalf of the Israeli military and facilitating the killings of … ⌘ Read more

@prologic@twtxt.net I couldn’t find the exact blog post from before, one that used redirection directives in its nginx config. but I found [this one ](https://melkat.blog/p/unsafe-pricing#:~:text=Something%20else%20I’ve%20been%20doing%20this%20year,%20fine.) mentioning a similar process but done differently.

Investigation into pre-Budget leaks is under way, MPs told
Chancellor Rachel Reeves has given her “full support” to a review of security processes around the lead-up to the Budget. ⌘ Read more

Here’s what experts say about the Russia–Ukraine peace process
After a flurry of meetings, Russia’s war in Ukraine continues. So where is the peace process at? ⌘ Read more

Sound Open Firmware 2.14 Released With Intel Wildcat Lake & Nova Lake Support
Sound Open Firmware is one of the projects started originally by Intel but has grown into a multi-vendor initiative for open-source audio digital signal processing (DSP) firmware and development tooling for a variety of platforms under the Linux Foundation umbrella… ⌘ Read more

openSUSE Begins Rolling Out Intel NPU Support
Via the openSUSE Innovator Initiative, packaging of the Intel Neural Processing Unit (NPU) driver for the openSUSE ecosystem has begun. This is helping to jump-start the Intel NPU support within the openSUSE space although user-space applications ready to leverage the Intel NPU still remains very limited… ⌘ Read more

Colleges Are Preparing To Self-Lobotomize
The skills that future graduates will most need in an age of automation – creative thinking, critical analysis, the capacity to learn new things – are precisely those that a growing body of research suggests may be eroded by inserting AI into the educational process, yet universities across the United States are now racing to embed the technology into every dimension of their curricula.

O … ⌘ Read more

‘Kangaroo court’: UK MP lashes Bangladeshi court over corruption conviction
Tulip Siddiq says the decision should be treated with “contempt”, while her political party says she was not afforded due process. ⌘ Read more

Multiple meat workers in hospital after carbon monoxide leak in Melbourne
More than a dozen workers have fallen ill after a carbon monoxide leak at a pork processing plant in Derrimut. ⌘ Read more

(cont.) CM continua a desmanchar o discurso de AV (ou vai fazendo entre interrupções). Está bem, claro, e meteu AV nervoso, mas pena é que já perdemos 14 minutos e ainda não se fez nada a não ser falsr do que o AV quiz trazer para aqui

1. Patriarca de LX critica a normalização do discurso de ódio - AV não vê nada de mal com o discurso de ódio e volta a falar daquilo que quiz logo desde o minuto zero. Catarina Martins diz que obviamente tem razão o discurso de ódio ser um problema, o RASI confirma, há uma ligação directa com a extrema direita. AV chama censura à regulação das redes sociais, e que a CMartins não pode dizer que defende as mulheres porque está do lado de Gaza. CMartins fala de censura quando AV processou CMartins (e perdeu) por ela ter dito que ele era racista (que é) num debate. AV começa na berraria, Gaza e burcas e cenas, mas CMartins explica como defende os direitos humanos. Mais berraria e tal.
   
2. Jornalista pergunta a AV mais uma coisa, AV em vez de responder diz várias coisas que são más (segundo ele) em CM para ver se ela não tem tempo para corrigir cada uma delas. Vendo que não tem tempo, CM fala antes do problema das notícias falsas, e pede que as pessoas tenham cuidado e não caiam em notícias falsas (em sequência das mentiras de AV)

Why Google’s custom AI chips are shaking up the tech industry
Google is reportedly in talks to sell its tensor processing units – a type of computer chip specially designed for AI – to other tech companies, a move that could unsettle the dominant chip-maker Nvidia ⌘ Read more

Controversy over City of Perth’s $125,000 workplace culture review
Internal revolt intensifies at the City of Perth over the newly elected lord mayor’s pursuit of a ratepayer-funded workplace culture review, in a process deemed unlawful by his own administration. ⌘ Read more

cat /etc/mokou/yarnd.conf
exec=/usr/pkg/sbin/daemonize -c/var/db/yarnd -u www -p /var/run/yarnd.pid /usr/pkg/sbin/chpst -e /usr/local/etc/yarnd /usr/local/sbin/yarnd -b 127.0.0.1:[classified information]

I know this might seem a bit overengineered, but the previous command until now had the secrets exposed on the process list

@prologic@twtxt.net The periodic blacklists updates will be done automatically in the background, as for the different processing mechanisms (rules, collections of rules, remediation …etc) you just install/add the pre-made ones from the hub and call it a day, they’ll get periodic updates when needed. But you could easily create and add your own in case you want to block or white-list a specific behavior

@prologic@twtxt.net The main thing that I tought of is that whomever is abusing your services must be a well known actor (by range/set of IPs) that got reported by other Crowdsec users. So to my simpleton’s understanding, your reverse-proxy/web server passes the requests by crowdsec for processing, they get banned for $N hours if the source has already been blacklisted by the community or violates any of a set of behavior base rules (and even more hours for repeat offenders); otherwise the requests/responses go as per usual. Not sure if I got things right but this might help paint a better picture of the process.

Breaking: US stops processing immigration requests from Afghan nationals
The decision comes hours after US President Donald Trump said he was confident the suspect behind the shooting of two National Guard soldiers in Washington was from Afghanistan. ⌘ Read more

We may need a fourth law of thermodynamics for living systems
The laws of thermodynamics don’t accurately account for the complex processes in living cells – do we need a new one to accurately measure the ways living systems are out of equilibrium? ⌘ Read more

Nvidia Claims ‘Generation Ahead’ Advantage After $200 Billion Sell-off on Google Fears
Nvidia pushed back against investor concerns about Google’s competitive positioning in AI on Tuesday after the chipmaker’s shares tumbled 4.4% and erased nearly $200 billion in market cap on fears that Alphabet’s tensor processing units were gaining ground against its dominance in AI computing. The company said … ⌘ Read more

ProcessOne: Stop Telling Us XMPP Should Use JSON

Image

We hear this too often: “XMPP uses XML. It should use JSON—it’s more modern.”

The logic seems straightforward: JSON came later, so it must be better. But better for what, exactly?

JSON became successful because it’s the standard serialization format for JavaScript. That made it convenient for browser-based applications.

Does that m … ⌘ Read more

CDC Changes Webpage To Say Vaccines May Cause Autism, Revising Prior Language
A Centers for Disease Control and Prevention webpage that previously made the case that vaccines don’t cause autism now says they might. WSJ: The contents of the webpage came up during Health Secretary Robert F. Kennedy Jr. Senate confirmation process. Sen. Bill Cassidy (R., La.) in February said Kennedy had assured hi … ⌘ Read more

Updated Steam Runtime Switches To Debian 13 Libraries, SDL2 Using Compatibility Layer
An updated version of the Steam Linux Runtime 4 branch was rolled out that has now shifted from Debian 11 to Debian 13 libraries for some significant upgrades. In the process more libraries have gone x86_64 only in foregoing the i386 builds. In addition, the SDL 2 library support for the Steam Runtime is now provided by sdl2-compat as the compatibility layer for SDL2 atop SDL3… ⌘ Read more

Ultra-Processed Food is Global Health Threat, Researchers Warn
Action is needed now to reduce ultra-processed food (UPF) in diets worldwide because of their threat to health, say international experts in a global review of research. From a report: They say the way we eat is changing - with a move away from fresh, whole foods to cheap, highly-processed meals - which is increasing our risk of a range of chronic … ⌘ Read more

Iran Begins Cloud Seeding To Induce Rain Amid Historic Drought
Authorities in Iran have sprayed clouds with chemicals to induce rain, in an attempt to combat the country’s worst drought in decades. From a report: Known as cloud-seeding, the process was conducted over the Urmia lake basin on Saturday, Iran’s official news agency Irna reported. Urmia is Iran’s largest lake, but has largely dried out leaving a vast … ⌘ Read more

Waveshare Pairs RISC-V ESP32-P4 and ESP32-C6 for Wi-Fi 6, Bluetooth 5 LE, and PoE Support
Waveshare has released the ESP32-P4-WIFI6-POE-ETH, a compact development board built around the ESP32-P4 along with an ESP32-C6 wireless module. The design combines Wi-Fi 6, Bluetooth 5 LE, Ethernet, and optional PoE power delivery in a single platform aimed at multimedia processing, display and camera applications, and general embedded development. Like the earlier W … ⌘ Read more

OAK 4 D and OAK 4 S Standalone Edge Vision Cameras with PoE and 48MP Imaging
Luxonis has opened early access preorders for the OAK 4 D and OAK 4 S, two standalone edge-processing cameras designed for computer vision tasks. Both systems provide a 48MP RGB sensor with optional autofocus or wide-angle variants, USB 3 and PoE connectivity, IP67-rated enclosures, and on-device inference capabilities. Both devices are built around the RVC4 […] ⌘ Read more

Be it Java with Swing or PyQt6, it takes ~300 ms until a basic window with a treeview and a listbox appears. That is a very noticeable delay.

Is it unrealistic to expect faster startup times these days? 🤔

Once the program is running, a new second window (in the same process) appears very quickly. So it’s all just the initialization stuff that takes so long. I could, of course, do what “fat” programs have done for ages: Pre-launch the process during boot, windowless. But I was hoping that this wasn’t needed. 😞 (And it’s a bad model anyway. When the main process crashes, all windows crash with it.)

Apple Cuts App Store Fee In Half For ‘Mini Apps’
Apple is cutting its App Store fee from 30% to 15% for developers who join a new Mini Apps Partner Program, which requires using more of Apple’s built-in technology to power lightweight “mini apps.” “This includes using Apple software to register a user’s purchase history, verify user ages and to process in-app purchases,” reports CNBC. From the report: A “mini app” is a lightwei … ⌘ Read more

Russia’s AI Robot Falls Seconds After Being Unveiled
Russia’s first AI humanoid robot, Aldol, fell just seconds after its debut at a technology event in Moscow on Tuesday. “The robot was being led on stage to the soundtrack from the film ‘Rocky,’ before it suddenly lost its balance and fell,” reports the BBC. “Assistants could then be seen scrambling to cover it with a cloth – which ended up tangling in the process.” … ⌘ Read more

OpenAI’s GPT-5.1 Brings Smarter Reasoning and More Personality Presets To ChatGPT
OpenAI today released GPT-5.1, an update to its flagship model line. The update includes two versions: GPT-5.1 Instant, which OpenAI says adds adaptive reasoning capabilities and improved instruction following, and GPT-5.1 Thinking, which adjusts its processing time based on query complexity.

The Thinking model respond … ⌘ Read more

FFmpeg To Google: Fund Us or Stop Sending Bugs
FFmpeg, the open source multimedia framework that powers video processing in Google Chrome, Firefox, YouTube and other major platforms, has called on Google to either fund the project or stop burdening its volunteer maintainers with security vulnerabilities found by the company’s AI tools. The maintainers patched a bug that Google’s AI agent discovered in code for decoding a 1995 vi … ⌘ Read more

New Linux Patches To Expose AMD Ryzen AI NPU Power Metrics
New Linux kernel patches currently undergoing review will allow AMD Ryzen AI NPU power metrics to be exposed under Linux. In turn this is useful for helping to gauge the utilization of the neural processing unit and also helping to evaluate the actual power efficiency of leveraging the AMD Ryzen AI NPU… ⌘ Read more

Engicam Showcases Computer Vision AI Kit Based on Renesas RZ/V2H Platform
Engicam has showcased the TIA RZ/V2H System-on-Module and its companion AI.DEV RZ/V2H development kit, both built around Renesas’ RZ/V2H processor. The module targets embedded applications in machine vision, autonomous robotics, and industrial automation, offering onboard AI acceleration and GPU capabilities optimized for real-time processing and advanced imaging. The TIA RZ/V2H module integrates … ⌘ Read more

A universal law explains the chaotic motion of chromosomes
Researchers from Skoltech, the University of Potsdam, and the Massachusetts Institute of Technology have discovered a fundamental physical law that governs the seemingly chaotic motion of chromosomes inside a living cell. This discovery helps solve a long-standing biological mystery of how two-meter-long DNA molecules, packed into dense chromosomes, remain mobile enough for vital processes such as turning genes on and off. ⌘ Read more

Falling asleep isn’t a gradual process – it happens all of a sudden
Brain activity from more than 1000 people shows a rapid transition from being awake to being asleep, rather than a slow transition between the two states ⌘ Read more

ProcessOne: On Signal Protocol and Post-Quantum Ratchets

Image

Signal improved its protocol to prepare encrypted messaging for the quantum era.

They call the improvement “Triple Ratchet” (or SPQR = Signal Post-Quantum Ratchet).

[Signal Protocol and Post-Quantum Ratchets\
\
We are excited to announce a significant advancement in the security … ⌘ Read more

@bender@twtxt.net The only problem with uploading is the procesing. Do you expect any server-side processing of the WebP or just store and host?

@prologic@twtxt.net Let’s go through it one by one. Here’s a wall of text that took me over 1.5 hours to write.

This section says AI should not be treated as an authority. This is actually just what I said, except the AI phrased/framed it like it was a counter-argument.

The AI also said that users must develop “AI literacy”, again phrasing/framing it like a counter-argument. Well, that is also just what I said. I said you should treat AI output like a random blog and you should verify the sources, yadda yadda. That is “AI literacy”, isn’t it?

My text went one step further, though: I said that when you take this requirement of “AI literacy” into account, you basically end up with a fancy search engine, with extra overhead that costs time. The AI missed/ignored this in its reply.

Okay, so, the AI also said that you should use AI tools just for drafting and brainstorming. Granted, a very rough draft of something will probably be doable. But then you have to diligently verify every little detail of this draft – okay, fine, a draft is a draft, it’s fine if it contains errors. The thing is, though, that you really must do this verification. And I claim that many people will not do it, because AI outputs look sooooo convincing, they don’t feel like a draft that needs editing.

Can you, as an expert, still use an AI draft as a basis/foundation? Yeah, probably. But here’s the kicker: You did not create that draft. You were not involved in the “thought process” behind it. When you, a human being, make a draft, you often think something like: “Okay, I want to draw a picture of a landscape and there’s going to be a little house, but for now, I’ll just put in a rough sketch of the house and add the details later.” You are aware of what you left out. When the AI did the draft, you are not aware of what’s missing – even more so when every AI output already looks like a final product. For me, personally, this makes it much harder and slower to verify such a draft, and I mentioned this in my text.

You, @prologic@twtxt.net, also mentioned this in your car tyre example.

In my text, I gave two analogies: The gym analogy and the Google Translate analogy. Your car tyre example falls in the same category, but Gemini’s calculator example is different (and, again, gaslight-y, see below).

What I meant in my text: A person wants to be a programmer. To me, a programmer is a person who writes code, understands code, maintains code, writes documentation, and so on. In your example, a person who changes a car tyre would be a mechanic. Now, if you use AI to write the code and documentation for you, are you still a programmer? If you have no understanding of said code, are you a programmer? A person who does not know how to change a car tyre, is that still a mechanic?

No, you’re something else. You should not be hired as a programmer or a mechanic.

Yes, that is “skill evolution” – which is pretty much my point! But the AI framed it like a counter-argument. It didn’t understand my text.

(But what if that’s our future? What if all programming will look like that in some years? I claim: It’s not possible. If you don’t know how to program, then you don’t know how to read/understand code written by an AI. You are something else, but you’re not a programmer. It might be valid to be something else – but that wasn’t my point, my point was that you’re not a bloody programmer.)

Gemini’s calculator example is garbage, I think. Crunching numbers and doing mathematics (i.e., “complex problem-solving”) are two different things. Just because you now have a calculator, doesn’t mean it’ll free you up to do mathematical proofs or whatever.

What would have worked is this: Let’s say you’re an accountant and you sum up spendings. Without a calculator, this takes a lot of time and is error prone. But when you have one, you can work faster. But once again, there’s a little gaslight-y detail: A calculator is correct. Yes, it could have “bugs” (hello Intel FDIV), but its design actually properly calculates numbers. AI, on the other hand, does not understand a thing (our current AI, that is), it’s just a statistical model. So, this modified example (“accountant with a calculator”) would actually have to be phrased like this: Suppose there’s an accountant and you give her a magic box that spits out the correct result in, what, I don’t know, 70-90% of the time. The accountant couldn’t rely on this box now, could she? She’d either have to double-check everything or accept possibly wrong results. And that is how I feel like when I work with AI tools.

Gemini has no idea that its calculator example doesn’t make sense. It just spits out some generic “argument” that it picked up on some website.

The AI makes two points here. The first one, I might actually agree with (“bad bot behavior is not the fault of AI itself”).

The second point is, once again, gaslighting, because it is phrased/framed like a counter-argument. It implies that I said something which I didn’t. Like the AI, I said that you would have to adjust the copyright law! At the same time, the AI answer didn’t even question whether it’s okay to break the current law or not. It just said “lol yeah, change the laws”. (I wonder in what way the laws would have to be changed in the AI’s “opinion”, because some of these changes could kill some business opportunities – or the laws would have to have special AI clauses that only benefit the AI techbros. But I digress, that wasn’t part of Gemini’s answer.)

Except for one point, I don’t accept any of Gemini’s “criticism”. It didn’t pick up on lots of details, ignored arguments, and I can just instinctively tell that this thing does not understand anything it wrote (which is correct, it’s just a statistical model).

And it framed everything like a counter-argument, while actually repeating what I said. That’s gaslighting: When Alice says “the sky is blue” and Bob replies with “why do you say the sky is purple?!”

But it sure looks convincing, doesn’t it?

This took so much of my time. I won’t do this again. 😂

Tape containing UNIX v4 found
A unique and very important find at the University of Utah: while cleaning out some storage rooms, the staff at the university discovered a tape containing a copy of UNIX v4 from Bell Labs. At this time, no complete copies are known to exist, and as such, this could be a crucial find for the archaeology of early UNIX. The tape in question will be sent to the Computer History Museum for further handling, where bitsavers.org will conduct the recovery process. I have the equ … ⌘ Read more

Open Container Initiative “OCI” Runtime Spec v1.3 Released With FreeBSD Support
The Open Container Initiative unveiled today the OCI Runtime Specification v1.3 update for this standard around operating system process and application containers. This runtime specification continues to evolve for outlining the configuration, execution environment, and lifecycle of a container. Notable with the v1.3 revision is introducing official FreeBSD support… ⌘ Read more

Black Sea nitrous oxide conundrum: Why most N₂O produced by microbes never reaches the surface
Microorganisms in the Black Sea can produce large amounts of the potent greenhouse gas nitrous oxide (N2O). However, this gas never reaches the atmosphere because it is swiftly consumed by other microorganisms, which convert it to harmless dinitrogen gas (N2). Scientists from the Max Planck Institute for Marine Microbiology have now investigated this process and identified the key players involved. ⌘ Read more

ProcessOne: Europe’s Decentralized Messaging Survives “Chat Control” Threat

Image

Good news for anyone building messaging infrastructure in Europe: Denmark&aposs Council presidency is abandoning mandatory detection orders in the Child Sexual Abuse Material (CSAM) proposal for now. The proposal was nickna … ⌘ Read more

Microsoft breaks Task Manager in Windows 11, hard
Let’s take a look at how things are going at Microsoft, whose CEO claimed a few months ago that 30% of their code was generated by “AI”. After installing Windows Updates released on or after October 28, 2025 (KB5067036), you might encounter an issue where closing Task Manager using the Close (X) button does not fully terminate the process. When you reopen Task Manager, the previous instance continues running in the background even th … ⌘ Read more

Rate my AI teacher? Students’ perceptions of chatbots will influence how they learn with AI
A “transformation” is upon us. After a multi-year procession of educational technology products that once promised to shake things up, now it’s AI’s turn. ⌘ Read more

ProcessOne: AI Bots Can’t Use WhatsApp Anymore. So… Who Are They Going to Talk To?

Image

Meta just closed the gates on AI chatbots. I think this is an early warning.

Starting January 15, 2026, [WhatsApp will ban all third-party general-purpose AI chatbots from its platform](https://techcrunch.com/2025/10/18 … ⌘ Read more

@movq@www.uninformativ.de Yeah, give it a shot. At worst you know that you have to continue your quest. :-)

Fun fact, during a semester break I was actually a little bored, so I just started reading the Qt documentation. I didn’t plan on using Qt for anything, though. I only looked at the docs because they were on my bucket list for some reason. Qt was probably recommended to me and coming from KDE myself, that was motivation enough to look at the docs just for fun.

The more I read, the more hooked I got. The documentation was extremely well written, something I’ve never seen before. The structure was very well thought out and I got the impression that I understood what the people thought when they actually designed Qt.

A few days in I decided to actually give it a real try. Having never done anything in C++ before, I quickly realized that this endeavor won’t succeed. I simply couldn’t get it going. But I found the Qt bindings for Python, so that was a new boost. And quickly after, I discovered that there were even KDE bindings for Python in my package manager, so I immediately switched to them as that integrated into my KDE desktop even nicer.

I used the Python KDE bindings for one larger project, a planning software for a summer camp that we used several years. It’s main feature was to see who is available to do an activity. In the past, that was done on a large sheet of paper, but people got assigned two activities at the same time or weren’t assigned at all. So, by showing people in yellow (free), green (one activity assigned) and red (overbooked), this sped up and improved the planning process.

Another core feature was to generate personalized time tables (just like back in school) and a dedicated view for the morning meeting on site.

It was extended over the years with all sorts of stuff. E.g. I then implemented a warning if all the custodians of an activitiy with kids were underage to satisfy new the guidelines that there should be somebody of age.

Just before the pandemic I started to even add support for personalized live views on phones or tablets during the planning process (with web sockets, though). This way, people could see their own schedule or independently check at which day an activity takes place etc. For these side quests, they don’t have to check the large matrix on the projector. But the project died there.

Here’s a screenshot from one of the main views:

This Python+Qt rewrite replaced and improved the Java+Swing predecessor.

Magnetic gel could remove kidney stones more effectively
Standard techniques for removing kidney stones often require repeated surgery, but a magnetic gel seems to make the process more efficient ⌘ Read more

Electrochemical system converts plant compound into two valuable products at once
A research team has created a new kind of electrochemical “two-in-one” system that turns plant-based molecules into two useful products at once. Using a finely tuned single-atom ruthenium catalyst, the process combines two chemical reactions, oxidation and hydrogenation, inside a single electrolytic cell, much like cooking two dishes in the same pot without mixing up the flavors. Details of the research were p … ⌘ Read more

Quantum-inspired algorithm could help reveal hidden cosmic objects
Combining a quantum-inspired algorithm and quantum information processing technologies could enable researchers to measure masses of cosmic objects that bend light almost imperceptibly ⌘ Read more

ProcessOne: 🚀 ejabberd 25.10

Image

Release Highlights:

• Added more Ad-Hoc Commands from XEP-0133
  
• Updated support for XEP-0317 Hats
  

If you are upgrading from a previous version, there are no mandatory changes in SQL schemas, configuration, API commands or hooks.

Other contents:

• **[New option archive_muc_as_mucsub in mod_mam]( … ⌘ Read more