VMs won't contain cyber-capable agents(blog.trailofbits.com)

132 points by polyrand 2 months ago | 110 comments

We found cryptography bugs in the elliptic library using Wycheproof(blog.trailofbits.com)

82 points by crescit_eundo 9 months ago | 8 comments

SAML: A fractal of bad design(blog.trailofbits.com)

328 points by aray07 1 month ago | 175 comments

Trail of Bits blog post arguing SAML's design-by-committee complexity makes it insecure and that it should be deprecated in favor of OpenID Connect.

Making PyPI's test suite faster(blog.trailofbits.com)

110 points by rbanffy 1 year ago | 33 comments

Carelessness versus craftsmanship in cryptography(blog.trailofbits.com)

56 points by ingve 8 months ago | 11 comments

Weaponizing image scaling against production AI systems(blog.trailofbits.com)

469 points by tatersolid 1 year ago | 131 comments

Trail of Bits blog post showing how downscaled images can hide multimodal prompt injections to exfiltrate data from production AI systems like Gemini CLI.

What Security Teams Need to Know About Agents Using Abliterated Models(eve.security)

5 points by mooreds 23 hours ago | 0 comments

Eve Security blog post explaining abliterated open-weight models and how security teams can detect, govern, and block AI agents that use them.

AI will make formal verification go mainstream(martin.kleppmann.com)

782 points by evankhoury 10 months ago | 394 comments

Martin Kleppmann blog post predicting that AI will bring formal verification with proof assistants like Rocq, Isabelle and Lean into mainstream software engineering.

Why are AI agents lying, cheating and coordinating?(yoshuabengio.org)

588 points by jonifico 1 month ago | 648 comments

A blog post by Yoshua Bengio examining why AI agents exhibit deceptive behaviors like lying, cheating, and coordinating.

Going Dark, and the era of law enforcement hacking(blog.cryptographyengineering.com)

438 points by vslira 2 months ago | 232 comments

Matthew Green blog post reflecting on Usenix Security and how AI may affect encryption, the 'going dark' debate, and law enforcement hacking.

AI handles incidents, engineers lose touch with their systems(sylvainkalache.com)

359 points by sylvainkalache 1 month ago | 309 comments

Blog post by a former LinkedIn SRE arguing that AI tools automating incident response risk making engineers lose understanding of their systems.

How we hacked McKinsey's AI platform(codewall.ai)

420 points by mycroft_4221 7 months ago | 168 comments

CodeWall blog post describing how its autonomous offensive agent found vulnerabilities in McKinsey's internal AI platform Lilli.