Remediating Vulnerabilities With LLMs: Inside Ivanti's Automation Push
Summary
Ivanti CSO Daniel Spicer says frontier models have shown surprising effectiveness in early stages; but cost and human-in-the-loop viability remain open questions.
Discussion on
Trending posts from X.
“What happens if OpenAI/anthropic does this?” used to be the #1 question I got from vcs (which is silly because anthropic is not going to enter the reinsurance business, and if they do, we would win). Now with open source models at the frontier and more choices than ever, it…
— nico laqua (@nico_laqua) July 21, 2026
Sakana AI just launched a multi-agent system built specifically for cybersecurity...
— Wes Roth (@WesRoth) July 21, 2026
and it is already competing with the strongest specialized frontier models.
Fugu-Cyber scored 86.9% on CyberGym, where agents have to examine real codebases and verify actual vulnerabilities.… https://t.co/8OBqH82TeV pic.twitter.com/Jz39kop0wK