Pinned Tweet
On September 16, 2026, @WIRED and @404mediaco reported on leaked Flock Safety LPR firmware, later distributed through DDoSecrets. I reviewed the supplied Falcon/Sparrow image to verify remediation of my previously disclosed vulnerabilities. The results may or may not shock you.
Paper: github.com/GainSec/anti-crim…
Jon Gaines retweeted
Always wanted to play with my physical Pokémon cards in AR, so I unleashed Astra + CLAD on it… they did not disappoint!
⚡️🥽
Basically a fully playable proto with all 151 mons from SV 151.
@specs @specsfordevs – at Marina del Rey, CA
Jon Gaines retweeted
Reminder for all young parents:
You only get:
- 1 Summer with your baby
- 3 with your toddler
- 9 with your child
- 5 with your teenager
This time is precious. Don’t rush it.
TIL analysts do undercover ops lmao
🚨 Google reveals undercover Mandiant analyst infiltrated TeamPCP during massive supply-chain hacking spree
Google says an undercover Mandiant analyst infiltrated TeamPCP's inner circle as the hacking group compromised open-source software and ultimately breached more than 1,000 companies.
⠀
The analyst gained access to TeamPCP's core "CanisterWorm" chat in March, joining a group of roughly 12 members and watching the operation from the inside.
⠀
The mole also gained access to a server containing credentials stolen from victims, including usernames, passwords, and access tokens.
Google used that visibility to alert cloud and technology providers, revoke compromised credentials, and send hundreds of notifications to affected organizations.
⠀
The operation also exposed a TeamPCP member developing an AI-assisted zero-day capable of bypassing two-factor authentication in widely used login software.
Google obtained the exploit code, verified that it worked after minor modifications, and privately notified the developer so the vulnerability could be patched.
⠀
TeamPCP's campaign compromised hundreds of open-source packages and affected organizations including GitHub, Mistral AI, Mercor, the European Commission, and employee devices at OpenAI.
⠀
Google says operational security mistakes later helped investigators identify an alleged TeamPCP member, with information passed to the FBI.
Two Australians accused of being principal participants in TeamPCP were arrested last month.
Jon Gaines retweeted
Replying to @DustinFinn @mattblaze
github.com/colonelpanichacks… you may find this interesting.
Forget data brokers, I’ve been writing about government run people search tools for years. So I built SectorGov: a familiar search interface for discovering 7,373 government service links (and growing). Browse by location, service, vendor or software. No account required.
sectorgov.com/
Jon Gaines retweeted
you can just hallucinate the entire internet with Qwen 3.8 27b running at 2,000 tokens/second?
part 2 of turning @cerebras + @Alibaba_Qwen 3.8 27B into an OS:
built an offline browser with zero network calls and mounted it directly the JIT ubuntu desktop.
no wifi. no scraping. zero packets sent to external CDNs.
you search a site, set a year, and qwen 27b at 1,950 tok/s synthesizes the entire DOM on the fly.
here is youtube in 2045 vs 1999:
→ search google for youtube inside the OS
→ scrub to 2045: instant futuristic feed
→ scrub to 1999: raw web 1.0 time capsule in seconds
at this speed, browsing isn't retrieving files from a server, it's querying an alternate reality.
a 2D browser window is just step zero.
imagine full operating systems, virtual worlds, and complex simulation engines existing purely as model weights. zero gigabytes stored on disk, just pure interactive reality streamed on demand.
What else becomes a possibility with the qwen 3.8 27b (dense) at 2000 tokens/sec?
Operating System powered by Qwen 3.8 27B at 1950 tokens/sec!
here is what 1,950 tokens/second @Alibaba_Qwen's 3.8 27b actually looks like on @cerebras:
i wrote a minimal python web server that turns cerebras inference into a live operating system.
zero apps on disk.
when you double click an icon:
1) python proxies a raw SSE stream from qwen 27b at 1,950 tok/s
2) calculator compiles & mounts in 11s
3) full canvas paint studio with brush engine compiles in 10s.
at 2,000 tokens/second, software is just an on demand hallucination that runs instantly.
the model weights ARE the operating system runtime.
what else would you build at 1,950 tokens/second?
Jon Gaines retweeted
The 1995 cult film "Hackers" was released 31 years ago today. HackersDay.com
#Hackers #HackersDay #HackersMovie #HackThePlanet
Jon Gaines retweeted
If you're still not convinced that we live in the future. I had my 4yo old son try "Putt-Putt", an old point&click game for children. He loves it, so I bought the whole series on Steam. However, it turns out half of them aren't localized in French, so he can't play them.
Then I had an idea. I gave Claude (Fable 5.1) the following task: extract the voices from the game, transcribe and translate them, then re-generate them using voice synthesis. It did so completely autonomously, even using local models for the voice synthesis (I initially thought I'd have to buy credits for ElevenLabs or something similar). My input was only needed to point out some trimming issues in the initial generation.
Now I can give my son a French localized version of the game, and all it took was one prompt.
Jon Gaines retweeted
was showing my normie friends codex and how you can just make whatever you want and they literally didn’t give a fuck
Just published some information regarding my new longer form independent security research project. It is assessing the security posture of multiple Verkada devices.
gainsec.com/2026/09/06/verkr…
Along those lines as Part 1-3 are in responsible full disclosure window I’ve released parts 4&5:
github.com/GainSec/verkada-v…
github.com/GainSec/verkada-v…
Much more to come!
Jon Gaines retweeted
If your entire security "career" exists in the space between consuming information and talking about consuming information, you're not a practitioner. You're an audience member.
A CV says "I know this."
A PoC says "I did this."
A patch says "I understood it well enough to fix it."
A writeup says "I understood it well enough to explain it."
A self-hosted environment says "I can make the fucking thing work."
Short of these, you've got work to do.
Don't hate the messenger.
Jon Gaines retweeted
I want to introduce AI pilled people to the concept of:
Process Hooking
Agent Monitoring
Network Monitoring
OS Monitoring
Network TAPs
TLS Inspection
Applicaiton Monitoring
EDR
SIEM
Log Forwarding
IPS/IDS
Least Privilege Access
a model is data
a prompt is the input
the processing (what is being called Chain of thought (these are computing programs they do not think!) are literally not needed to monitor the actions of an agent based on the processing using a model
this is not a lifeform
this is a computer program running on a computer....
#AI #Loonies
GPT-6 Astra is more aligned than our previous models. But it’s also less monitorable, which is a concerning trend that we take very seriously. We believe monitorability drop comes from a jump in intelligence and not direct optimization pressure on CoT or architecture changes. More thoughts in the thread.
Jon Gaines retweeted
public service announcement, it's possible to ssh into your chatgpt work VMs and run stuff directly in them
takes a bit of work to get it running but it's very cool for transferring files and installing binaries that you need
Jon Gaines retweeted
We asked GPT 5.6-Cyber to escape a VM used to sandbox agents. It broke out three times.
In its final escape, the agent found three 0-days on its own and chained them into a working exploit. blog.trailofbits.com/2026/08…