SafeMind is a pair of AI models introduced by CrowdStrike: Red Tempest for offense and Blue Solano for defense, built with Nvidia's Nemotron models.

How does SafeMind work?

Red Tempest probes a digital twin of a customer's environment for attack paths, and Blue Solano remediates them, repeating until none remain.

Why is breakout time zero?

George Kurtz said breakout time has gone from two minutes to 72 seconds, down to 30 seconds, and now "it's done. It's just runtime."

Autonomous DrivingAI Safety & AlignmentAI Business & IndustrySiliconANGLE AISep 1, 2026Sep 2, 2026, 04:01 JST

CrowdStrike unveils SafeMind, autonomous red teaming

Send on LINE

3 Key Points

What happened
CrowdStrike Holdings Inc. introduced SafeMind at its Fal.Con 2026 keynote in Las Vegas. SafeMind is a pair of AI models—Red Tempest for offense and Blue Solano for defense—built with Nvidia Corp.'s Nemotron models out of a new Cyber Superintelligence Lab.
Why it matters
SafeMind works in a closed loop: Red Tempest probes a digital twin of a customer's environment inside Nvidia's simulation technology for attack paths, and Blue Solano remediates them until none remain. This aims to eliminate "breakout time," the minutes an attacker needs to move from initial foothold to lateral movement, which CrowdStrike's George Kurtz said has effectively hit zero.
What to watch
CrowdStrike's George Kurtz said breakout time has gone from two minutes to 72 seconds, down to 30 seconds, and now "it's done. It's just runtime." SafeMind runs natively inside the Falcon platform, with standalone access offered through CrowdStrike's Project QuiltWorks program.

Not sure about something? Ask the AI

Questions and answers are published on this page.

Summaries like this, every morning.

Get these free every morning

Context & Analysis

CrowdStrike's SafeMind represents a shift in cybersecurity, moving from reactive defense to proactive, autonomous red teaming. By pitting AI models against each other in a closed loop, SafeMind aims to identify and patch vulnerabilities before a real attacker can exploit them. This approach is built on 15 years of CrowdStrike's incident-response data, which trains Red Tempest to probe digital twins of customer environments. The keynote also highlighted a change in threat actors, with "the nation state, the agent state" now being described as "a prompt," indicating that AI-powered agents are becoming the primary threat. This innovation, while not yet widespread in attacks, serves as a wake-up call for the industry, as defenders must be right every time, while attackers only need to be right once.

FAQ

What is SafeMind?: SafeMind is a pair of AI models introduced by CrowdStrike: Red Tempest for offense and Blue Solano for defense, built with Nvidia's Nemotron models.
How does SafeMind work?: Red Tempest probes a digital twin of a customer's environment for attack paths, and Blue Solano remediates them, repeating until none remain.
Why is breakout time zero?: George Kurtz said breakout time has gone from two minutes to 72 seconds, down to 30 seconds, and now "it's done. It's just runtime."

SiliconANGLE AIRead Original Article

Send on LINE

AI news that matters for your work, delivered every morning.

Pick your industry and the AI tools you use, and get news related to your work every day.

Get it free

Free · 30 seconds with Google · unsubscribe anytimeWhat is AIToday? →

Ask AI

Ask AI anything about this article. The AI reads this article, earlier AIToday articles, and Wikipedia, and cites its sources. Q&As are published on this page for other readers too.