The security officer emphasized that organizations need to identify weak points and verify their defense capabilities before adversaries take advantage of vulnerabilities.

Key points:

• At the AI Proving Grounds Consortium ( AIPGC ) event held on September 24th, Gartner ® Vice President Analyst Tom Powledge and cybersecurity leaders from SimSpace, Corelight, and Dreadnode discussed why attacks driven by AI are accelerating the shift towards preventive security.
• As the time for AI to perform compression detection and respond to attacks increases, organizations need to identify potential attack paths and verify their defenses before their opponents can take advantage of them.
• Realistic network simulations and continuous validation can help organizations test security controls, while also demonstrating the resilience of critical business operations.

Orlando, Florida, November 12 ( PRNewswire ) – At the event " Build Trust in AI : Govern , Measure and Scale " held on September 24, organized by AI Proving Grounds Consortium ( AIPGC ), Vice President Analyst Tom Powledge of Gartner along with the heads of Corelight , Dreadnode , and SimSpace discussed why attacks driven by AI require a shift in security strategies towards proactive security. AIPGC brings together cybersecurity leaders to help organizations securely train, test, and validate AI defenses before production deployment.

The discussion explored how AI can accelerate network attacks and shorten the response time of defenders, thereby highlighting the importance of realistic attack simulations and continuous security validation, in order to identify vulnerabilities before attackers can exploit them.

“This is true. We have already begun to see these attacks,” said Powledge.

Powledge indicates that as AI accelerates the discovery of vulnerabilities and the actions of adversaries, organizations need to 'shift to a more proactive, preventive, and structured approach,' and increasingly adopt methods such as continuous threat exposure management, adversarial validation, and simulation. 'We want to take action before attacks occur,' he said.

As the response window approaches zero

As the time between intrusion and the onset of impacts continues to shorten, this challenge becomes even more urgent.

SimSpace Global Live CISO and Chief Evangelist Jason Rivera pointed out that the time it takes for attackers to breach security, that is, the time between the initial access and lateral movement, is significantly decreasing, and speculated on what would happen when this window approaches zero. "This is where the concept of the 'Proactive Security Era' comes from," Rivera said.

Defenders do not have to wait for the opponent to attack the production system; instead, they can use realistic simulations of their own environment to identify potential attack paths, test defenses, and establish protective measures before such attacks occur.

The goal is to regain the initiative. “What if you could launch various attacks against yourself before your opponent does, thereby truly taking the lead in stopping them?” said Rivera.

Assume that an intrusion has already occurred. Prevent attacks in advance. Demonstrate resilience.

Proactive security does not assume that you can prevent every attack. It assumes that a determined adversary may ultimately gain access, and it questions whether the organization has proven that its most important systems, personnel, security controls, and AI proxies are capable of withstanding what might happen next. This changes the ultimate measure of network security readiness.

The question is no longer just: How fast can we respond? But rather: When we are invaded or our systems are compromised, can the most critical business functions and tasks still maintain resilience?

Continuous security testing is still far from being widespread. According to the "2026 State of Agentic Cybersecurity" report published by SimSpace, only 29% of the surveyed organizations conduct continuous network simulations, which means that many organizations have limited opportunities to verify their defenses in response to current threats.

AI Proving Grounds provides a realistic environment where organizations can securely test simulated versions of their own infrastructure without compromising the production systems. This allows for the identification of weak points, verification of AI proxies and security controls, training of human operators, and continuous improvement of defense measures.

Organizations can take advantage of this environment to:

• Discover the attack path before the opponent takes advantage of it.
• Verify whether the AI proxy, security controls, and human operators are functioning as expected.
• Prioritize and address weak links based on the impact of business and tasks.
• Generate new defense capabilities through realistic attack simulations.
• Prove continuously that critical systems can maintain resilience in the face of threats and environmental changes.

As both attackers and defenders begin to operate at machine speed, the advantage in security may increasingly belong to those organizations that are able to anticipate the next attack before it actually occurs.

This discussion reaffirmed the focus of AIPGC: to help organizations assess the security capabilities driven by AI under realistic conditions, and to enhance their confidence in defending against emerging threats.

View the discussion on AIPGC.

The complete “Build Trust in AI : Govern , Measure and at Scale” event is available for viewing as needed. It includes Gartner, the Vice President Analyst; Tom Powledge; Corelight’s Greg Bell; Dreadnode’s Brad Palm; and SimSpace’s Jason Rivera. https :// aipgc.ai / aipgc-events /

Organizations that wish to apply these principles can also understand how AI Proving Grounds helps teams to build, verify, and deploy AI defenses under realistic adversarial conditions. https :// simspace.com / ai-proving-grounds /

Frequently Asked Questions

What is preventive network security? Proactive network security is a method of moving security activities to an earlier stage of the attack lifecycle. Rather than relying solely on detection and response after the adversary takes action, organizations also use simulation, exposure validation, threat intelligence, and continuous testing to identify potential attack paths and strengthen defenses before attacks reach the production environment.

Why is AI changing the network security response time? AI can increase the speed, scale, complexity, and persistence of cyberattacks. As the time between initial access and lateral movement is reduced, defenders may not have much time for manual investigation and response. This makes preventive testing and automated defense increasingly important.

What is AI Proving Grounds? AI Proving Grounds is a realistic environment where organizations can safely build, train, test, validate, and deploy AI agents, security technologies, and manual work processes under adversarial conditions, before applying them in production. It allows organizations to simulate attacks that are realistic versions of those targeting their own environments, identify exploitable attack vectors, test security controls and AI agents, generate new defenses, and continuously retest as threats and systems evolve.

Will preventive security replace incident response? No. Detection, threat hunting, and incident response are still of paramount importance. Proactive security adds an extra layer of protection at an earlier stage, helping organizations to identify and prepare before adversaries can exploit potential attack vectors.

About AI Proving Grounds Consortium

AI Proving Grounds Consortium brings together leaders in network security to help organizations securely train, test, and validate AI defenses before production deployment. Founding members include Corelight, Dropzone AI, SCYTHE, SimSpace, and Sondera. For more information, see AIPGC.ai.

Media Contact

Amy Rice PR for SimSpace [ email protected ]

Gartner is a trademark of Gartner, Inc and/or its affiliated companies.