#claude#claudecode#vulnerabilityresearch
12K subscribers • 1 video on Hallway
Advanced. Deep dive for people who know the field.
388K views7 months ago10 recommendations
SecurityAI and machine learningTransformers and LLMsExploits and vulnerabilitiesWeb and application security
Nicholas Carlini explains how modern large language models can automate the discovery and exploitation of zero-day vulnerabilities in heavily tested software.
Nicholas Carlini, Research Scientist, Anthropic, speaks at [un]prompted 2026 on: Black-hat LLMs. Large language models are now capable of automating attacks that were previously only possible by human adversaries. In this talk, I discuss several ways that adversaries could mis-use current models in order to cause harm both at a larger scale and at a lower cost than they do currently. For example, we find that recent state-of-the-art models can now find 0-day vulnerabilities in large software projects that have been extensively tested by humans for decades. These new capabilities will alter the threat landscape and require we rethink security in the coming years.What people say. Recommended as an eye-opening demonstration of how advanced LLMs have become at finding and exploiting real-world security vulnerabilities.
knowaveragejoeon Hacker News • Recommends
On “Project Glasswing: Securing critical software for the AI era”
panstromekon Hacker News • Recommends
On “Claude Wrote a Full FreeBSD Remote Kernel RCE with Root Shell (CVE-2026-4747)”
ogigon Hacker News • Recommends
On “Claude Wrote a Full FreeBSD Remote Kernel RCE with Root Shell (CVE-2026-4747)”
eichinon Hacker News • Recommends
On “Claude Code Found a Linux Vulnerability Hidden for 23 Years”
woeiruaon Hacker News • Recommends
On “Project Glasswing: Securing critical software for the AI era”
conradkayon Hacker News • Recommends
On “Project Glasswing: Securing critical software for the AI era”
underdeserveron Hacker News • Recommends
On “Project Glasswing: Securing critical software for the AI era”
vmg12on Hacker News • Recommends
On “Small models also found the vulnerabilities that Mythos found”
wkandekon Hacker News • Recommends
On “Project Glasswing: what Mythos showed us”
tptacekon Hacker News • Mentions
On “Vulnerability Research Is Cooked”
jsnellon Hacker News • Mentions
On “Claude Wrote a Full FreeBSD Remote Kernel RCE with Root Shell (CVE-2026-4747)”
59nadiron Hacker News • Mentions
On “The Claude Code Leak”
SyneRyderon Hacker News • Mentions
On “OpenAI Acquires TBPN”
e12eon Hacker News • Mentions
On “Claude Code Found a Linux Vulnerability Hidden for 23 Years”
klausaon Hacker News • Mentions
On “NSA is using Anthropic's Mythos despite blacklist”