GLM 5.3 can autonomously compromise recent versions of Mozilla Firefox
Summary
Epoch AI 发布研究报告指出，GLM 5.3 能够自主攻破 Mozilla Firefox 的近期版本，揭示了开放权重模型在补丁差距（patch gap）上的网络安全风险。
Similar Articles
AI can turn security fixes into working exploits before users receive an update (16 minute read)
Epoch AI 研究发现，开源模型 GLM 5.3 能够利用安全补丁发布与用户更新之间的'patch gap'，自主构建针对 Firefox 的零点击文件窃取漏洞利用，成本仅约 1,000 美元/次，表明广泛可用的开源模型已能独立完成此前需要专家级能力的漏洞利用开发。
Defending the patch gap against automated attackers
Mozilla explains how AI models like GLM 5.3 can accelerate patch-gapping attacks and outlines Firefox's new defenses, including faster weekly releases, holding back regression tests, commit scanning, and a structural defense-in-depth approach.
We have a year to fix security everywhere
The article discusses the release of GLM 5.3-flash, an open-weight AI model that is cheap and capable, raising concerns about potential misuse for hacking. It calls for urgent action to fix security vulnerabilities across the tech industry using frontier LLMs.
@seclink: Zhipu AI (https://Z.ai) today released GLM-5.3, which shares the same base model as GLM-5.2, with all improvements from post-training reinforcement learning (RL). 【1】Programming: Strongest in open-source, but still behind closed-source frontiers GLM-5.3 achieved...
Zhipu AI released GLM-5.3, significantly enhancing programming and cybersecurity capabilities through post-training reinforcement learning, becoming the top open-source model for programming, and unexpectedly discovering numerous real vulnerabilities.
Watch as a vulnerability cybersecurity expert with 10 years of experience in the field becomes astonished by the reverse engineering powers of... GLM 5.3
A low-level vulnerability researcher with 10 years of experience discovered that using GLM 5.3 and Ghidra MCP, firmware extraction, decryption, and verification of a Dell monitor — work that would normally take about a week — was completed in just 16 minutes, raising concerns about the "commodification" of reverse engineering capabilities by AI.