Being able to find security vulnerabilities in publicly available models is the real flex nowadays.
That’s really separates the average security researcher from the elite ones.
Introducing GPT‑6 Cyber: a major advance in security research that pairs stronger capabilities with better model alignment.
The model powered the Linux kernel research we announced yesterday, and security teams at companies like Mozilla, Palo Alto Networks, Calif, and Cisco—alongside our own at OpenAI—are already putting it to work.
GPT-6 Cyber is available to select Daybreak Red customers and we are expanding access over time.
https://t.co/Nzab3Z1Td6
We have identified strong indications that the employee device reportedly involved in the recent Binance security incident is the same device we detected during the final days of June 2026.
Our investigation revealed that the device was infected with #RisePro Infostealer* on **June 23, 2026**, potentially exposing a significant volume of sensitive credentials, authentication tokens, and cloud access keys.
Compromised Data Identified:
- 6 Google OAuth Tokens
- 2 AWS Secret Access Keys
- 2 AWS Access Key IDs
- 2 macOS Keychain artifacts
- 4 AWS Key Pairs
- Google OAuth session artifacts
Additional Exfiltrated Data:
- 548 Credentials
- 865 Cookies
- 1,882 Autofill Records
- 12 Secrets
do you know how strong exploit mitigations in modern operating systems are? what a modern exploit chain looks like? or how many vulnerabilities/bugs it can take to turn a memory corruption bug into code execution.
just shut up and patch ur system.
thanks
Opened an old phone from pre-2022 and found a screenshots of exploits/code/projects
This job used to be pure dopamine. Nothing felt better than popping up a calculator
Miss these days when human critical thinking was more important than LLM reasoning/efforts level
2026 experience
> sign up to ovh
please upload your passport
> sign up to latitude sh
pay 25$ OR upload your passport
> sign up to hetzner
> omg it works, buy server
[Ticket#123456711] RE: Manual order review
im buying ssh access to a computer man not controlled substances
Today, we're kicking off the first phase of the research preview for Model Hardware Standard (MHS): a new standard for AI agents to safely operate physical equipment in scientific research and advanced manufacturing.
Read more: https://t.co/XQ2y9EW7Af