Emergent misalignment is strongest in GPT-4o and Qwen2.5-Coder-32B-Instruct, but all fine-tuned models exhibit inconsistent behavior https://t.co/J7aXekrw1D
"This technique enables hackers to silently compromise AI-generated code by injecting hidden malicious instructions into seemingly innocent config files used by Cursor and Copilot" 😟https://t.co/U8typIF187
A look at four ways that AI is reshaping hacking and malware development, and how we can stay vigilant in response.
@stacklokhq
https://t.co/9BMlCtP9Bh
🎉 Welcome to the OpenSSF family, Minder!
📣 @StackLokHQ is contributing Minder to OpenSSF as a sandbox project! Minder streamlines #OSSSecurity, auto-remediates issues, and flags key risks for devs & security teams.
🔍 Learn more about Minder: https://t.co/i1Z4Ee2HY2
“90% of the code that's being delivered into a production environment is written by random people on the internet. And those random people are increasingly using generative AI models”—@cmcluck, on Stacklok donating its Minder #security supply chain platform to @openssf#ATO2024
Join Stacker @jaosorior for an Intro to #Minder today at 10am ET / 3pm BST / 5pm EEST to get a high-level overview and demo of the project.
YouTube livestream is at: https://t.co/BQCiOSPTag
#SupplyChainSecurity#ShiftLeft#DevSecOps
Now available in #Minder: Profile Selectors give you the flexibility to customize how & when policies are applied to your projects. Easily customize how Minder profiles are applied to your projects, and apply the right rules to the right resources.
https://t.co/AtNsq5uwSH
📣TODAY! Join @puerco at #SOSSCommunity Day Europe as he dives into how to build a trusted end-to-end VEX stream, from code to scanner diving deep into a VEX document and explores other highlights of the OpenVEX ecosystem.
Thu. Sep 19 at 10:40am CEST
https://t.co/RkqAUASuXS