The cybersecurity market is approaching a moment when artificial intelligence could alter the balance between attackers and defenders. As increasingly capable models become available, the question facing security leaders is no longer whether AI will influence cyber defense, but how quickly organizations can deploy it in ways that produce measurable security outcomes.
OpenAI is now calling for that transition to accelerate. In its new open letter, “A Call to Action on Cyber Defense,” the company is urging cybersecurity vendors, governments, technology companies and organizations to coordinate around AI-powered defense. Palo Alto Networks, Mate Security, and ServiceNow are among the first-day signatories to the initiative, putting three companies with very different approaches to security on the same list of market leaders supporting the effort.
For Mate Security, the announcement carries particular significance. The company is among a relatively small group of cybersecurity providers building specifically around agentic cyber defense, an emerging category in which AI systems can reason through investigations and execute security workflows rather than simply assist human analysts.
A Race Between Offense And Defense
OpenAI’s letter frames the issue as a race against time.
“We have a limited window to strengthen cyber defenses,” it states.
The company expects AI-enabled attacks to become more widespread and sophisticated as models around the world improve. That could put additional pressure on organizations already struggling with vulnerabilities that have accumulated over decades.
The list is familiar to security executives: unpatched systems, excessive permissions, insecure software, weak authentication, misconfigurations and legacy technical debt. The difference now is that AI may provide defenders with a way to address those weaknesses at a scale that traditional security teams cannot easily match.
OpenAI argues that “AI brings specialist skills to more defenders and makes core security tasks faster, cheaper and better.”
Different Generations Of Cybersecurity
The participation of the three companies illustrates how broad the AI transition is becoming.
Palo Alto Networks represents the established cybersecurity market, with AI increasingly incorporated into a large security ecosystem. ServiceNow represents the enterprise platforms where security work is assigned, tracked and closed out across teams. Mate Security, by contrast, is building around a newer model in which autonomous agents become active participants in security investigations.
The distinctions may become increasingly important as security operations evolve.
For years, automation has largely meant translating human-defined rules into software workflows. Agentic systems introduce a different possibility: software that can interpret evidence, determine the next investigative step and adapt as circumstances change.
Mate Security’s Gamebooks approach is designed around this model. The technology provides AI agents with structured investigation procedures while allowing them to reason, pivot and act as new information emerges.
The remediation side matters to the same equation. OpenAI asks organizations to fix their most dangerous weaknesses and confirm that those fixes work, which turns on how reliably that work moves through the systems where it is coordinated. Faster investigation delivers less than expected if the response behind it still runs at manual speed.
The Trust Challenge
The shift toward autonomous defense also creates a problem that OpenAI highlights in its recommendations: organizations need to trust what AI agents do.
OpenAI calls on frontier AI companies to build observability and security tools and ensure that “agentic identities are traceable and accountable.”
That requirement is particularly relevant as enterprises consider allowing AI systems to move from analyzing security events to taking action. Autonomous defense could significantly expand the capacity of security teams, but only if organizations can establish clear boundaries, accountability and verification.
The letter therefore focuses heavily on practical deployment rather than AI experimentation. OpenAI calls for organizations to fix their highest-risk weaknesses, verify the results and build security into what they purchase, develop and deploy, including AI-generated code.
A Broader Industry Commitment
The initiative also rejects the idea that one technology company or cybersecurity vendor can independently solve the emerging problem.
“No single company should control the future,” OpenAI says, arguing that cyber capabilities are advancing globally and that the response must therefore be global as well.
That means greater sharing of threat intelligence, tested playbooks and verified fixes, along with additional resources for critical infrastructure organizations that may not have the budgets to deploy advanced defensive technologies themselves.
For Palo Alto Networks, Mate Security and ServiceNow, becoming first-day signatories represents an endorsement of that collective approach.
The more significant development, however, may be what their participation says about the cybersecurity market itself. AI-powered defense is increasingly moving from an experimental category into a strategic priority, with established platforms, enterprise workflow providers and agentic-native companies all positioning themselves to help organizations navigate the next generation of cyber threats.
John Kevin Hao is a news and feature writer covering cybersecurity, technology, and business targeted for professional audiences.