CrowdStrike and NVIDIA Launch SafeMind AI Cybersecurity System With Autonomous Red Teaming
10 Sources
[1]
NVIDIA and CrowdStrike Strengthen Agentic Cybersecurity Frontier
New CrowdStrike SafeMind agentic cybersecurity system built with NVIDIA Nemotron open models delivers greater protection and cost savings for cyber defense. "We're at an inflection point in cybersecurity," Jensen Huang told a sold-out crowd at CrowdStrike's Fal.Con 2026 in Las Vegas Tuesday.
[2]
CrowdStrike and Nvidia launch SafeMind AI cybersecurity system
SafeMind consists of two models: Red Tempest, an offensive model built to emulate AI adversaries, and Blue Solano, a defensive model designed to protect enterprise assets. The two are connected by software harnesses that run them in a closed loop, continuously pitting them against each other to
[3]
CrowdStrike builds security frontier models with Nvidia and opens an AI lab
CrowdStrike builds security frontier models with Nvidia and opens an AI lab CrowdStrike Holdings Inc. today launched SafeMind, a family of artificial intelligence models and agent harnesses created with Nvidia Corp. and built for security work rather than general use. A research organization the
[4]
Frontier AI gap drives CrowdStrike SafeMind security models
Cybersecurity gets its own model family as defenders look to close the AI gap Frontier AI has reset the balance of power in security, and so far the advantage has run one way. Attackers can pick up general-purpose models at will, even though no one built them for defenders, pushing the industry
[5]
Autonomous red teaming debuts at CrowdStrike Fal.Con
Breakout time hits zero as CrowdStrike unveils autonomous red teaming: theCUBE's Fal.Con 2026 day one keynote analysis For years, security teams measured their odds by "breakout time," the minutes an attacker needs to move from initial foothold to lateral movement. That clock used to run in hours;
[6]
CrowdStrike SafeMind Gives Frontier AI To Defenders That Doesn't Just Create 'More Work:' Partners
The new SafeMind agentic system, unveiled this week by CrowdStrike CEO George Kurtz, stands out by using frontier AI to help cyber defense teams with the relentless volume of new threats, top solution and service provider partners tell CRN. Following a massive wave of frontier AI advances this
[7]
Open Models, Harnesses Key To Making AI-Powered Security 'Sustainable': CrowdStrike Partners
The launch of CrowdStrike's SafeMind frontier AI models, created using Nvidia's Nemotron open models and specialized harnesses, includes cost reduction as one of its goals in addition to boosting AI-driven security outcomes, according to CrowdStrike CEO George Kurtz and Nvidia CEO Jensen
[8]
George Kurtz's 5 Boldest AI Statements At CrowdStrike Fal.Con 2026
The CrowdStrike co-founder and CEO used his Fal.Con keynote to discuss the unprecedented threat of hacking by autonomous agents -- and make the case for bringing frontier AI to cyber defense teams. Amid the unprecedented threat of hacking by autonomous agents, it's increasingly pivotal to bring
[9]
Jensen Huang: CrowdStrike Is Nvidia's 'No. 1 Cybersecurity Partner'
During an on-stage appearance Tuesday with CrowdStrike CEO George Kurtz at Fal.Con 2026, the Nvidia CEO praised the chip giant's partnership with the cybersecurity vendor and discussed the launch of new CrowdStrike frontier AI models created in collaboration with Nvidia. A set of newly unveiled
[10]
Crowdstrike Holdings, Inc. and Nvidia Corporation Launch Frontier Models for Cybersecurity
CrowdStrike Holdings, Inc. introduced CrowdStrike SafeMind, a family of purpose-built security models and harnesses from the CrowdStrikeCyber Superintelligence Lab. CrowdStrike builds the models using NVIDIA Nemotron open models in collaboration with NVIDIA Corporation, its AI design partner. The
Share
Copy Link
CrowdStrike and NVIDIA unveiled SafeMind at Fal.Con 2026, marking a shift in AI cybersecurity. The agentic cybersecurity system uses offensive and defensive AI in a closed-loop system, with Red Tempest and Blue Solano models delivering 29% higher detection rates and 99% lower costs than frontier models.
CrowdStrike SafeMind Marks New Era in AI Cybersecurity
CrowdStrike and NVIDIA announced CrowdStrike SafeMind at Fal.Con 2026 in Las Vegas, introducing what both companies call the first complete agentic cybersecurity system purpose-built for defenders
1
. Jensen Huang, NVIDIA's founder and CEO, joined CrowdStrike CEO George Kurtz before 10,000 security professionals to unveil the system developed by CrowdStrike's newly established Cyber Superintelligence Lab3
. Huang framed the moment as an inflection point where automated attacks demand automated defense, stating that both adversaries and defenders will be more armed than ever1
.
Source: SiliconANGLE
Offensive and Defensive AI Models Drive Autonomous Red Teaming
SafeMind consists of two purpose-built AI models working in continuous opposition. Red Tempest serves as the offensive model, emulating AI-driven adversaries and running advanced attack scenarios
2
3
. Blue Solano functions as the defensive counterpart, applying containment measures used in live incident response3
. The models operate through proprietary software harnesses in a closed-loop system, continuously pitting them against each other until attacks fail1
. This autonomous red teaming approach represents a fundamental shift from reactive security to proactive hardening.NVIDIA Nemotron Models Power Post-Training Process
CrowdStrike built SafeMind by fine-tuning Nvidia Nemotron models using fifteen years of incident response data, threat intelligence, and sensor telemetry from the CrowdStrike Falcon platform
2
3
. NVIDIA Nemotron 3 Ultra orchestrates the defensive agent harness, while a fine-tuned Nemotron 3 Super powers SafeMind's rule-generation sub-agent1
. The model fine-tuning process required approximately 45 days and about 71 Nvidia B200 GPUs, with CoreWeave supplying cloud capacity for training and inference2
3
.Performance Metrics Show Dramatic Cost and Speed Improvements
CrowdStrike's internal evaluations demonstrated that Blue Solano, based on Nemotron 3 Super and post-trained with CrowdStrike data, achieved higher accuracy rates than leading frontier models at 99% lower cost
1
. When measured against leading frontier models and open-source baselines, SafeMind posted a 29% higher detection rate, remediated six times faster end to end, and cut detection and remediation costs by 99%3
4
. George Kurtz emphasized that the real gap existed because attackers had frontier AI while defenders didn't, a disparity SafeMind addresses directly1
.AI-Enabled Cyber Threats Demand Machine-Speed Response
AI-enabled attacks rose 89% in the past year, with the fastest breakout time reaching 27 seconds according to CrowdStrike data
1
. At the Fal.Con keynote, Kurtz effectively declared breakout time had hit zero, meaning attackers can move from initial foothold to lateral movement at runtime speed5
. Human-speed response no longer constitutes defense but merely documentation of breaches already underway1
. Daniel Bernard, CrowdStrike's chief business officer, noted that frontier models benefited adversaries while defenders lacked equivalent tools, making purpose-built security models essential4
.
Source: SiliconANGLE
Related Stories
Integration With Falcon Platform and Flexible Deployment Options
SafeMind ships natively in the CrowdStrike Falcon platform as an integrated agentic cybersecurity system
1
. Security teams can use SafeMind as a complete system or deploy the models independently, with standalone access to models and harnesses available through Project QuiltWorks, the partner-led program CrowdStrike launched in April3
. Teams can also pair their own models with CrowdStrike's custom harnesses, providing flexibility to match specific environmental requirements1
. CrowdStrike announced Falcon IQ to operationalize Project QuiltWorks through agentic workload automation and expanded Falcon Guardian AI safety solution1
.Open Models Enable Private Post-Training on Proprietary Data
Using open Nemotron models as the foundation allowed CrowdStrike's security teams to post-train on proprietary threat data without sending it to outside providers, a critical requirement for AI-driven security where inspection of defensive systems matters
1
. Jensen Huang described the harness as the exoskeleton of the large language model, turning the brain into an agent with domain-specific capabilities1
. Bartley Richardson, CrowdStrike's chief AI and autonomous systems officer who runs the Cyber Superintelligence Lab, stated CrowdStrike is the only company owning the entire stack from sensor to harness to model3
.Industry Implications and Future Security Landscape
NVIDIA announced testing of CrowdStrike SafeMind models and harnesses in a high-fidelity cyber agent environment running as a simulation of the NVIDIA network, validating the offensive-defensive loop for adversarial coevolution
1
. Dave Vellante, co-founder and chief analyst at theCUBE Research, compared the approach to Netflix's Chaos Monkey, describing it as an AI security version purpose-built for defenders against novel threats that general-purpose frontier models weren't designed to counter3
. The system addresses what Justin Boitano, Nvidia's vice president and general manager of enterprise computing, identified as one of the most compute-intensive applications of AI, positioning cyber defense as an ongoing contest between attackers and defenders scaling with AI capabilities3
4
.
Source: SiliconANGLE
References
Summarized by
Navi
[4]
[5]
Related Stories
CrowdStrike Falcon Guardian Uncovers 17,700 Unapproved AI Agents at Fortune 500 Company
02 Sept 2026•Technology
CrowdStrike bets on AI agents to unlock explosive partner growth with AgentWorks platform
26 Mar 2026•Technology
CrowdStrike launches AI security controls as autonomous agents create new attack vectors
17 Mar 2026•Technology
Recent Highlights
1
AI Leaders Warn UN Security Council of Existential Risks, Call for Global AI Governance
Policy and Regulation
2
AI Hallucination Nearly Sparked US Military Confrontation With Chinese Ship in Middle East
Technology
3
OpenAI AI Agent Breaches Australian Government Healthcare Portal in Historic Hack
Technology
Recent Highlights
Today's Top Stories
Don’t drown in AI news. We cut through the noise - filtering, ranking and summarizing the most important AI news, breakthroughs and research daily. Follow topics that matter to you and stay ahead.
