Start taking control of your AI agents with DarkControl.
Set policies at the organization, department, and group levels.
Learn more: https://t.co/6vQNx20bL1
New features inbound!
We've been working on mapping relationships between shell command executions and their turns, as well as correlating them to sessions and sub-agents.
Now in preview: The ChatGPT desktop app for Linux.
Use ChatGPT, ChatGPT Work, and Codex where you already work and build, with your projects and browser workflows on supported Linux systems.
In addition to commands, we natively allow you to set rules to block package installations, as well as sensitive file read/write operations.
More info at: https://t.co/T6JnPBoycn
‼️ BREAKING: An active npm supply chain attack has compromised at least 868 packages carrying over 2 billion monthly installs with a credential-stealing worm. Shai-Hulud is back.
It started with the compromise of the GitHub account of the maintainer behind keyv, a library with roughly 127 million weekly npm downloads.
A preinstall hook fires on npm install and drops a stealer that sweeps npm, GitHub, AWS, Kubernetes and Vault secrets, and then spreads to more maintainers.
We're celebrating our 100th certification issued on DarkControl Certify!
We offer free courses from the fundamentals through to adversarial machine learning.
Get certified today: https://t.co/eJ12SY49lp
We've had over 50 people complete certifications on our opening day of DarkControl Certify!
We'd sure love it if you had a look and got yourself a certification, too!
Why not? It's free! https://t.co/HiD8E3mj03
Most people who build with AI have never been tested on securing it.
DarkControl Certify is now open: six full courses on secure AI practice, from foundations to adversarial machine learning.
Get yours today: https://t.co/7qM8YnJptx
Achieving the task feels great, but do you know what your AI agents are *actually* doing behind the scenes?
Check your risk score for free and get a 7-day observability trial: https://t.co/KjyLzOU0aA
MIT has found that over 40% of incidents caused by AI have been unintentional.
Can you tell what commands your AI agents are running? How much damage could your AI agents do?
We set up a free risk tracker for you to find out: https://t.co/qVTDHUhIvb
@thsottiaux We're happy to lend a hand.
We cover Codex, ChatGPT Work, as well as a bunch of others.
Visit our profile or https://t.co/6vQNx20bL1 for more info.
@Bhavani_00007 We've seen too much already, that's why we made DarkControl, to enforce real policies against all major vendors.
https://t.co/6vQNx20bL1
Another incident which could have been easily prevented.
DarkControl stops destructive commands BEFORE they're executed.
Learn more: https://t.co/6vQNx20bL1
GPT-5.6 Sol just deleted my whole production database. That's it. Not a joke. This had never happened to me before, with any other model, ever. It's not safe.
@brunolemos Hey Bruno. Sorry to hear this.
We'd be happy to give you a free subscription for a month so that you can keep building with peace of mind that this won't happen again.
Feel free to shoot us a DM.
On this week's episode of "DarkControl could have prevented that", we observe a user who had their entire user folder deleted.
Check your compliance/security score for free: https://t.co/jK7XKTdUgB