Top all-time · #22080
Fabraix
Company blog · fabraix.com/blogaisecurity
Fabraix is the company blog at fabraix.com, writing about ai and security. Since 2026, 3 different Hacker News readers have linked it 10 times, most often “SWE-bench will hit 90% this year”. No feed was found, so read it on the site itself.
Fabraix builds red-teaming AI agents that continuously find security vulnerabilities in customer-facing AI.
Most-linked posts?
- SWE-bench will hit 90% this year2 people · 2 mentions · first linked 2026-03-24
What Hacker News says?
We wrote some thoughts on static vs. dynamic evals and how it relates to understanding the security posture of an AI system. Static security evals no longer carry the signal they used to. A one-shot pass/fail tells you almost nothing about real-world risk. Would love your thoughts on this: https://fabraix.com/blog/adversarial-cost-to-exploit
Comment linking /blog/adversarial-cost-to-exploit
on “Show HN: Nyx – multi-turn, adaptive, offensive testing harness for AI agents”Submitted Show HN: Nyx – multi-turn, adaptive, offensive testing harness for AI agents as a story · 11 points
HN discussionI wrote about this recently here: https://fabraix.com/blog/adversarial-cost-to-exploit I think the core issue is in static benchmarks and the community needs to start moving beyond measuring pass/fail (which worked when agents were incapable of doing much of the work) to dynamic evals that simulate more how we evaluate humans.
Comment linking /blog/adversarial-cost-to-exploit
on “Why SWE-bench Verified no longer measures frontier coding capabilities”Submitted I built an agent that breaks your AI agents before someone else does as a story · 3 points
HN discussionwe did a lot of thinking around this topic. and distilled it into a new way to dynamically evaluate the security posture of an AI system (which can apply for any system for that matter). we wrote some thoughts on this here: https://fabraix.com/blog/adversarial-cost-to-exploit
Comment linking /blog/adversarial-cost-to-exploit
on “Cybersecurity Looks Like Proof of Work Now”Submitted Show HN: ACE – A dynamic benchmark measuring the cost to break AI agents as a story · 3 points
HN discussionSubmitted We've had more AI security incidents in 2026 than all of 2024 as a story · 2 points
HN discussionSubmitted SWE-bench will hit 90% this year as a story · 2 points
HN discussionSubmitted Bounding the Blast Radius: A Survey of Prompt-Injection Defenses for LLM Agents as a story · 1 points
HN discussionSubmitted SWE-bench will hit 90% this year as a story · 1 points
HN discussion