
(#WinstonWolf + #BobSponge x Unicorn²) ^ #Cthulhu = Me ! #Cyberübersec #SysAdmin #DevOps #OpenBSD #HackThePlanet #HWPervert #Maker #infosec #Rawtherapee
$HOME=/tmp
Joined April 2010
- Tweets18.7K
- Following3.1K
- Followers797
- Likes23.1K
James L. retweeted
This is why computer vision is hard
Readers added context they thought people might want to know
The sprinkler activated due to heat from a real charcoal grill stove behind the sign at a barbecue stall (a customer was grilling skewers), not the flame graphic on it; the resort confirmed normal operation and adjusted the stove angle.
163.com/dy/article/LC4…
k.sina.cn/article_532885…
James L. retweeted
It has been discovered than you can control the M1 Tank with controller 2 in NG+ on MGS1. Looks like they worked on a 2 players feature for NG+ at the time and they changed their mind but forgot to remove the code. After almost 30 years, we keep discovering stuffs on this game 😅
James L. retweeted
David Kennedy -ancien hacker à la NSA- révèle qu'il travaille activement sur une cyberattaque en direct où un "essaim d'agents" d'IA -environ 30 agents- a contourné l'authentification à deux facteurs, volé des données et chiffré l'ensemble du système d'une entreprise : "Tout cela a pris 14 minutes. 14 minutes. Je n'avais jamais rien vu d'aussi rapide auparavant".
L'entreprise à mis plus de 4h30 a détecté l'intrusion... contre 14 minutes pour le piratage complet 🤯🤯🤯🥶
C'est plus hallucinant et inquiétant ce qui est entrain de se passer dans le monde avec l'exponentiation de l'IA et des essaims d'agents IA autonomes...
On se dirige vers le pire...
James L. retweeted
This is legit. GTA V Xbox build files circulating around are real.
I revise my original timeline for emulation to be <3 days (optimistic since we have yet to see XBOX emulation but I think this timeline is doable).
CYBERLEEK HAS LEAKED THE XBOX BUILD OF GTA 6🚨🚨🚨
They claim the full Xbox build (~208GB) is out via torrent, with the main executable locked behind a countdown, or released early if their crypto token hits a $15M market cap. Still unverified.
He claims the community will be able to emulate it within a week. If it’s real, we could see it running at high framerates on PC very soon!
James L. retweeted
Windows 11 26H2 lockscreen has been bypassed, this allows a full bitlocker bypass.
github.com/MSNightmare/FastU…
James L. retweeted
> be microsoft
> hire Indian CEO
> fires all the most loyal competent staff
> replaces them with Indians
> company goes to shit
As a Microsoft employee describes the situation..
James L. retweeted
> be NVIDIA
> world famous CEO
> Jensen Huang
> just a chill dude in a leather jacket
> has two affairs
> media outlets go bananas
> board asks Jensen Huang to resign
> Huang resigns
> says sorry, says "I'm a piece of shit"
> NVIDIA executives open dialogue with employees
> make suggestion box
> over 1,600 employees ask him to be reinstated
> "he's the boss"
> "we need him for the mission"
> "everyone makes mistakes"
> "i stand with Jensen"
> employees send out mass emails asking Jensen returns
> "NVIDIA isn't NVIDIA without Jenson"
> internal comms show overwhelming demand to reinstate Jensen
James L. retweeted
Wondering whether there's a pilot that has flown all the aircraft in the SR-XX series
James L. retweeted
🎂 Il y a désormais... 25 ans, je posais la première pierre de ce qui allait devenir Framasoft !
#TempusFugit
[tweet programmé] Il est 11h38 🎂 ;)
Il y a pile 20 ans naquit chez Gandi le 1er nom de domaine Framasoft.
Énorme merci aux personnes qui ont cru, soutenu, participé, un peu, beaucoup, passionnément à l'aventure.
Et à celles qui la poursuivent aujourd'hui.
soutenir.framasoft.org/fr/
James L. retweeted
Hacking a Biotronik cardiac pacemaker: Reverse engineering and exploitation of medical devices. ❤️🩹📟👨🏻💻🧲😈
More details on:
LinkedIn: lnkd.in/p/dVJjc_js
Substack: it4sec.substack.com/p/hackin…
Telegram: t.me/it4sec/114
James L. retweeted
The logic analyzer that Opus 5.5 laid out finally came back from JLCPCB! Time to put it through its paces!
James L. retweeted
I have a major result to share.
I’ve found conclusive forensic evidence that the top 4 Chinese AI labs—DeepSeek, Moonshot, MiniMax and Ant Group—have built their newest models on NVIDIA’s most advanced Blackwell chips, which are barred from export to China.
This sounds insane, but I discovered that almost every AI model carries a hidden signature of the GPU that trained it. When a model is first set up, the starting values of its weights are laid out in a pattern that depends on the exact GPU. Each combination of GPU and seed number produces a unique pattern—a sort of fingerprint of the hardware used to create the model. Traces of that pattern survive training into the published model.
Again, this evidence isn’t based on any insider information or leaking. It’s based on a forensic analysis of the model weights released by the companies themselves.
I searched more than six trillion combinations of chip layout and seed number and matched them to open-weights models. In doing so, I found conclusive evidence that six Chinese AI models—including DeepSeek-V4-Pro, Kimi K3, and MiniMax-M3—were initialized on NVIDIA Blackwell systems. Four of the models match Grace Blackwell, the GB200 and GB300 parts in NVIDIA’s flagship liquid-cooled rack-scale systems.
I then verified these results on real hardware. Running unmodified PyTorch with the seeds I’d recovered, a Blackwell B300 reproduced DeepSeek-V4-Flash’s untouched starting weights bit for bit, and a rented GB200 reproduced Kimi K3’s to a correlation of 0.999995. A Hopper H100 reproduced none of DeepSeek-V4-Flash’s, and changing the seed by one broke the match. (DeepSeek’s seed is 953049583 by the way.) The method also was verified to match 24 known open-weights models to the hardware that developers said they used.
It is exceedingly unlikely that these results are a coincidence. Across six trillion combinations, the best accidental match should reach only about 8 sigma (i.e., eight standard deviations above what random chance would produce). The real matches range from 49 to 1,385 sigma, and even the weakest single test hits 20.
These were data-center-scale training runs. With some analysis of release schedules and model details, I estimate that the largest, Kimi K3 and DeepSeek-V4-Pro, used 2,000 to 3,000 Blackwell GPUs each. Clusters of this size cost on the order of $150 to $200 million each to buy, and across these labs alone would add up to more than half a billion dollars of export-controlled GPUs, not counting experimental compute or otherwise. It’s quite likely that we see multi-billion dollar expenditure on export-controlled GPUs.
I am highly confident in this result, and it has been independently verified by senior researchers at OpenAI and Anthropic. You can read the full report at unemploymentlabs.com/checkpo….
There is much more I’ve identified about these models through other forms of checkpoint forensics like this, and will have more to share soon.
James L. retweeted
We have detected speed restrictions on Starshield on our Ukrainian territory – in the temporarily occupied parts of Ukraine. The speed limit is back at 161 kilometers per hour – that’s 100 miles per hour.
This directly affects our ability to intercept Russian jet-powered “shahed” drones. These drones come from those areas, and because of the speed restrictions, we simply cannot catch up with them in time.
Near the frontline, we cannot use our aircraft. Only drone interceptors can operate there. And without the speed we need, shooting down these “shaheds” becomes much harder.
This means more Russian strikes get through, and more of our people are killed by Russia. We are trying to get the details from the American side, but so far, we know very little.
If someone in America is making decisions that basically treat Ukrainian territory as if it is not Ukrainian, then this tells us where things are going. We need to be clear and never recognize Russia’s occupation of Ukrainian territory in any way.
From an address at the 9th European Political Community Summit. (3/3)
James L. retweeted
1/ AI is finding bugs faster than we can patch them, we hear "formally verify everything" more and more. So let’s look at the most verified kernel in existence, seL4, and what its latest release (Oct 30) fixed. 🧵
James L. retweeted
Still a favorite photo (@mikko and I at the Internet Archive, 2018)
James L. retweeted
The 3D codes that design fusion stellarators need exact solutions to check accuracy. Existing solutions covered at most 2 repeating segments, but real stellarators have 3 or more.
With Gemini 4 Carbon agents working together, we found solutions for any number N of segments.
James L. retweeted
Meu primeiro paper na @GoogleDeepMind
Colocamos dezenas de agentes do Gemini 4 Carbon para trabalhar nas equações diferenciais de dispositivos de fusão nuclear. Foram descobertas novas soluções exatas, incluindo uma que se sabia existir desde 1970, mas ninguém tinha encontrado.







































