Cohesity joins OpenAI's call for collective action on cyber defense

Attackers do not work in isolation. They reuse infrastructure, trade tools, and learn from successful attacks. AI is speeding up that process by reducing the time and expertise required to find vulnerabilities, craft convincing social engineering campaigns, and automate parts of an attack.

Defenders need to learn and move just as quickly. That is why Cohesity is joining OpenAI’s call for collective action on cyber defense alongside organizations from the technology, security, and critical infrastructure communities. The goal is straightforward: use AI to help defenders identify threats earlier, share what they learn, and better protect the organizations that depend on them.

Cohesity brings a specific point of view to this work. Prevention and detection are indispensable, but they are not enough. A sound cyber defense strategy must be built on a foundation of assured recovery.

That means being able to restore trusted data and critical systems quickly and safely when an attacker gets through. 

Recovery is foundational to security

Much of cybersecurity, quite reasonably, focuses on stopping attacks and finding adversaries that have entered an environment. AI can make both efforts faster and more effective.

But no CISO or CEO should build a security strategy on the assumption that every attack will be stopped. Ransomware operators continue to change their methods. Nation-state actors have significant time and resources. AI allows attackers of all skill levels to move faster.

Every organization needs to be ready for the possibility that one of those attacks will succeed.

At that point, the questions become highly practical. Can the company restore critical systems in the right order and get the business running again without reintroducing the threat? Can it trust its backup data? Can its security team determine whether that data contains malicious artifacts? 

That is cyber resilience in practice. Recovery is not an administrative task that begins after the security team has finished its work. It is an integral part of the defense strategy itself.

Secure AI depends on resilient data and infrastructure

AI raises the stakes because enterprise systems are no longer limited to analyzing information and offering recommendations. They are retrieving company data, supporting important decisions, and taking actions inside business systems.

If the underlying data has been manipulated, encrypted, or made unavailable, the consequences move quickly from an IT problem to a business problem. The same is true of the AI infrastructure itself. Agent configurations, memory, vector databases, and orchestration logic also need to be protected and recoverable. Operations can stop. Poor decisions can be based on unreliable information. Customers can be affected.

A company cannot trust its AI systems if it cannot trust the data and infrastructure behind them.

That is why secure AI requires resilient data and infrastructure. Organizations need to know that the information used by their AI systems is protected, properly governed, monitored for threats, and recoverable from a clean copy when something goes wrong.

This principle is at the heart of Cohesity’s Enterprise AI Resilience strategy. It also shapes what we hope to contribute through OpenAI’s call for collective action on cyber defense. 

What Cohesity brings to collective defense

Collective defense works when participants bring practical expertise and share what they are seeing across different parts of the threat landscape.

We combine research from our Cohesity REDLab with intelligence from strategic partners, government agencies, open sources, and customer-supplied indicators. That context helps organizations identify malicious artifacts within protected data before they begin a restoration, reducing the risk of bringing the threat back along with the data.

Signing OpenAI’s open letter gives us an opportunity to contribute our perspective on data protection and recovery. It also allows us to learn from researchers, technology providers, and infrastructure operators with different visibility into how threats are developing.

No company has a complete view of the threat landscape, and no single vendor can address every risk created by AI. Connecting the knowledge held across the industry gives defenders a better chance to recognize new techniques early and respond before they are widely used.

AI will make cyber defense faster and more effective. It will also allow attackers to operate at a speed and scale that security teams have not faced before. Organizations, therefore, need a strategy that covers the full course of an attack, including how they will restore trusted data and resume operations if their initial defenses fail.

By joining this call for collective action, Cohesity will continue making the case that recovery belongs in every cyber defense strategy from the start. Secure AI begins with data that can be trusted, protected, and restored.