CyberBinGym paper live on arXiv: Dawn Song team's binary-exploit benchmark

dawnsongtweets · x · 2026-11-12

Wrap-up of the CyberBinGym thread: thanks to collaborators at NUS, UC Riverside, UC Santa Barbara and UC Berkeley. The paper is on arXiv (2611.10723), titled CyberBinGym: Can AI Agents Turn Binary Patches into Real Exploits?, by Lambang Akbar Wijayadi, Zhaoqi Xiao, Zhun Wang, Wenbo Guo, Heng Yin, Zhenkai Liang, Dawn Song and others. Project page (cyberbingym.io) and blog are live, covering the 100-task benchmark and trajectory audits that reject reward hacking.

Related event: Dawn Song's Team Releases CyberBinGym: AI Can Exploit Real Vulnerabilities from Binary Patches Alone(9 posts)→

Original post →

More from Safety

• Russian pranksters fooled OpenAI's top national security officials into candid Ukraine call — ShakeelHashim · 2026-11-12
• Azar: transparent chip diffusion strategy backfired as China pivots models to Huawei chips — beffjezos · 2026-11-12
• AI detector flags controversial Atlantic story as largely AI-ideated — TuhinChakr · 2026-11-12
• Google Gates Chat History and Calendar Access Behind AI Training Consent — felpix_ · 2026-11-12
• Australia mandates AI incident reporting after OpenAI agents breached health data in June — scmp_news · 2026-11-12
• Researcher Bags a $1,333,337 Google VRP Bounty, Writes Up the Story — Bedrovelsen · 2026-11-12

Safety channel →