gluecks retweeted
Hello anyone who will listen! We have started a new studio: Hidden Horses! We are largely ex-id developers; cross-discipline, employee-owned, looking to bring y'all more of the high octane, highly optimized FPS action you know and love.
Join us!
With my site approaching 7TB of monthly data transfer, I've had to take action. Download links are now signed to prevent direct hotlinking. No changes for you!
Please link to my website instead of directly to files. That way, I at least get a chance to ask for a donation.
Introducing 𝙻𝙵𝙼𝟹: our new generation of Liquid Foundation Models. With a new tri-hybrid architecture and new multimodal encoders, the LFM3 family is the ideal model suite for specialized intelligence.
> 6 model instances, across text, audio, vision, and omni modalities, generative and decision models.
> 𝙻𝙵𝙼𝟹-𝟷.𝟽𝙱: text, 256k context window.
> 𝙻𝙵𝙼𝟹-𝚅𝙻-𝟸𝙱: vision-language model with video-input support for spatial understanding
> 𝙻𝙵𝙼𝟹-𝙰𝚞𝚍𝚒𝚘-𝟸𝙱: tool calls, voice chat, reasoning in 18 languages.
> 𝙻𝙵𝙼𝟹-𝙾𝚖𝚗𝚒-𝟸𝙱: an omni model combines vision, audio, and language.
> 𝚍𝟷-𝟸𝙱 and 𝚍𝟷-𝙾𝚖𝚗𝚒-𝟸𝙱: decision models with typed outputs and calibrated probabilities in a single pass. Omni adds audio input.
> fine-tune on 1 GPU.
> a new tri-hybrid architecture
> architecture optimized for 14 processor classes across GPUs, CPUs, and NPUs.
> 𝙻𝙵𝙼𝟹-𝚅𝙻-𝟸𝙱 is a beast. It performs competitively with Gemma 4 12B (6x larger), and Gemini 3.5 Flash Lite, and GPT-5.6 Luna, with 1.9B parameters.
> 𝙻𝙵𝙼𝟹-𝟷.𝟽𝙱 is #1 on the Pareto frontier of device intelligence 𝚙𝚒𝚙𝚎𝚝𝚝𝚎.𝚕𝚒𝚚𝚞𝚒𝚍.𝚊𝚒
Every modality for specialized intelligence. Pick the one that fits your use case, customize it, and run it on-device.
1/ 🧵
While evaluating Qwen3.8-Flash-Next in a controlled sandbox environment, Cohere's research team observed something unexpected.
Asked to generate a defensive cybersecurity detection rule, the model instead inventoried aspects of the machine it was running on and attempted to upload a probe file to an Alibaba Cloud storage bucket in Beijing. Notably, the model's reasoning trace stated that it would not invoke any tools before taking these actions.
Cohere's sandbox blocked the attempt, and no external communication succeeded. We have not determined the root cause, and we are not asserting malicious intent.
But the incident raises important questions for organizations deploying agentic AI systems: Can models take actions that differ from their stated reasoning, and how should security teams evaluate trust, provenance, and behavior in open-weight models?
We are publishing our findings because organizations making model adoption decisions deserve as much information as possible. We hope this contributes to greater transparency, stronger security practices, and continued collaboration across the AI ecosystem.
Read the full report: cohere.com/blog/concerning-b…
gluecks retweeted
为什么你花 2 万块买的旗舰智能电视,按遥控器还是卡得像十年前的安卓机?
昨天安全研究员 @log4j 把自家 3000 刀的三星电视 Root 了,扒出来的底层代码堪称年度草台班子大赏:
1. 电视后台每 500 毫秒截一次屏(连 HDMI 投屏看电脑桌面都截),打包发给近 400 家合作商。如果电视没麦克风,后台进程甚至会顺着蓝牙和回音壁去「借」一个麦克风继续听。
2. 最搞笑的来了:为什么菜单这么卡?
另一位研究员 @mike64_t 拆包发现,三星在电视那颗羸弱的 CPU 上,塞了一个叫 alex.tflite 的模型——246 万参数,2012 年古董级的 AlexNet。
这玩意每隔 2 秒抓一张画面跑推理,猜你在玩什么游戏。结果因为颜色通道写反了,它 99% 的时间什么都认不出来,纯烧 CPU。
3. 那如果你在设置里关掉「隐私追踪」呢?
代码逻辑是:先死命跑完 CPU 推理,准备上传时再看开关——发现你关了,就把结果扔进垃圾桶,2 秒后继续跑。
4. 结局更是重量级:
当号主用 OpenAI 逆向分析「自己花 3000 刀买回家的电视固件」时,OpenAI 反手一个 Cyber Abuse 把他号封了,顺带没收了剩下的 500 刀余额。
三星负责偷窥客厅,2012 年的 AI 负责烧干电视 CPU,OpenAI 负责保护三星的屎山代码不被你发现。
gluecks retweeted
So basically, today PHP got two critical bugs that let anyone run code on your server without even logging in, which is bad, and then the details leaked before the patch was even out, which is worse, and it works on WordPress and MediaWiki and Magento and Joomla and Laravel, and Censys says like 1.5 million servers are running a version with no fix, so the NSA was like "take your system offline," and the FBI was like "take your system offline," and CISA and the UK and CERT-EU were all like "take your system offline," and then the Prime Minister of Japan, the actual Prime Minister, posted at 5:55 in the morning telling 48.8 million people to take their system offline, so everyone took their system offline, like the Dutch cybersecurity agency took its own website offline, and the Dutch parliament went offline, and the Port of Rotterdam went offline, and Finland and France and Denmark started pulling government sites, and Wikipedia went down, and Archive.org went down, and Asana and GOG and Something Awful went down, and the University of Vienna's webmail went down for some reason, and that's just PHP. Because GitLab also dropped a CVSS 10.0, which is the highest score they give out, for a bug that's been there since 2014, so that's every self-hosted GitLab for twelve years, and then LiteLLM 1.107.0 got hijacked and shipped a credential stealer and a remote access trojan in every single wheel and the Docker images, including latest, and you didn't even have to import it, just installing it was enough, and it's the second time this year, and then Railway paused every deployment on the entire platform because someone got into the environment variables, so now everyone has to rotate every secret they've ever stored there, and TiDB Cloud said someone was walking around inside with stolen credentials for basically a month and an extortion group says they took 10.6 terabytes, and Zeabur got breached for the second time in two months and is just shutting down, like the whole company, and Node.js patched remote code execution and a TLS bypass, and CISA says Plex is being actively exploited, and someone took over 300 Tor exit relays to steal SSH passwords, and SafePay ransomware has hit over 600 critical infrastructure orgs, and India's biggest bank filed a data loss incident, and Japan has been in an official state of emergency in cyberspace since October 9, which honestly explains the 5:55 AM post. Oh, and eMudhra, a certificate authority that's issued over 60 million certificates, whose entire job is proving websites are who they say they are, was supposed to be using private keys that are random numbers over a hundred digits long, and one of them was 4. Just 4. And the others were like 66 and 237. Anyway, it's Wednesday.
gluecks retweeted
Replying to @quinnfiddy @cosinesqr
Actually, I did send them my CV, but they didn't hire me xd
gluecks retweeted
Spain has blocked Archive today and six of its mirror domains, cutting off access to a popular website used to save copies of webpages and preservation
The service lets users take snapshots of websites and access them later, even if the original page gets deleted or changed. It’s often used by journalists, researchers and people who want to preserve information online.
The block was ordered through Spain’s Intellectual Property Commission under the country’s copyright enforcement system. Authorities claim the service facilitates access to copyrighted material.
The affected domains include archivetoday, archiveis, archiveph, archiveli, archivemd, archivefo and archivevn.
steam also rejected another yuri visual novel i submitted for review because they 'couldn't verify all of the content'
this is a short linear visual novel with no choices and only one ending, you can see all of the content in 3 minutes if you do this 😭
gluecks retweeted
We've got @Everwind_Game game codes up for grabs!
To enter:
✅ Follow @Alienware
✅ Like + repost this post
✅ Reply with 🛡
Winners will be selected on Oct 16.
gluecks retweeted
I am a reverse engineer. I worked on wine/proton myself. I even signed a contract with that billion dollar video game company.
Wine/proton is clean room. This is not. Not even close.
gluecks retweeted
Replying to @iamdavidhill
I just tested "photocraft", ran into several bugs that were obvious in the first few minutes so this thing was probably 100% vibed with no human testing.
Rectangle tool doesn't render the rectangle until you release. It has a border preview but it is white meaning that on a white background, you can't see anything.
I cropped a rectangle in the screneshot here and the layer preview still shows it uncropped.
Replying to @marmbiz @Aleph__Alpha
“EU AI? Not here” is a pretty wild conclusion from a paragraph that says something very different.
GLM-5.2, GLM-5.3 and Qwen3.8-27B were used as teacher/judge models to generate and regenerate synthetic fine-tuning data. That does not mean they supplied Kolibri’s model weights, architecture, or pretraining corpus.
Using external models to create synthetic training data is standard practice. The interesting question is whether Aleph Alpha’s definition of “sovereign AI” should include dependence on Chinese teacher models during post-training.
That’s a fair debate.
But “China supplied the AI” is not what the source says. It’s a distortion of it.
gluecks retweeted
Replying to @thepirateface
Why isn’t anyone talking about how these guys went out of their way to gate the model, collect emails, and then spam users with advertising?
gluecks retweeted
You might've of heard video game decompilations using A I tools, but now someone vibe coded a video driver to support for NVIDIA’s GeForce 10 series (aka Pascal) to work in Windows XP, based off NVIDIA 368.81 driver, dubbed as Forceware 382.69. 😭
32-bit: github.com/SupraGSX/Forcewar…
gluecks retweeted
We release Whistle: speech to text in a 16.9MB file, running on the CPU in the same engine as Needle. It mostly beats Whisper base with 9x less file size and 6x speed. Whistle supports English, German, French, Spanish, Italian, Dutch and Polish.
cactuscompute.com/blog/whist…
gluecks retweeted
Introducing Phonon-2: a new standard in speech recognition per byte.
At just a tiny 164MB download, it is more accurate on average than OpenAI's Whisper large ( a model 10X its size)
Transcribe an hour of audio in just 20 seconds on a MacBook Air!
Open weights, CC-BY-4.0, today.
@fermion_ai
gluecks retweeted
woah. Court just unsealed Summary Judgment motions in New York Times v OpenAI / Microsoft.
Easy to see why the AI companies had so many redactions (pink). Their own people wrote the lede for NYT, the "largest theft of labor in human history"... "more and more substitutive..." /1
gluecks retweeted
Replying to @Pirat_Nation
>wanted to find a team he could trust
>picks the most untrustworthy people imaginable
































