Ukraine Is Turning to AI to Find the Hack Before Hackers Do

Hackers only need to find one weakness to break into a system. Ukraine’s cyber defenders, meanwhile, have to find those weaknesses first — and they are now getting help from an AI system built to do exactly that.

OpenAI is giving Ukraine’s Ministry of Digital Transformation access to Daybreak, its AI-powered vulnerability-hunting system, as the country faces a growing wave of cyberattacks targeting hospitals, power grids and government networks. The deployment puts OpenAI’s technology directly into the cyber defenses of a country already operating under sustained wartime pressure.

Daybreak is designed to search software for vulnerabilities, determine whether suspected weaknesses can actually be exploited and check whether a patch has successfully closed the hole. Rather than waiting for an attacker to discover a weakness, the system is built around finding it before that happens.

The rollout was announced Wednesday on the sidelines of the United Nations General Assembly by Dmytro Kushneruk, Ukraine’s consul general in San Francisco, and Sasha Baker, OpenAI’s head of national security policy.

“We are proud to support Ukraine’s cyber defenders, who are protecting essential services against attacks every day. Ukraine is already on the front line, and its defenders need support now.”

Ukraine Gives OpenAI’s Cyber Defense AI a Real-World Battlefield

The urgency behind the deployment is reflected in the scale of Ukraine’s cyber threat. CERT-UA, the country’s national computer emergency response team, handled nearly 6,000 cyberattacks in 2025, a 37% increase from the previous year, with local governments and state agencies among the hardest-hit targets.

For Ukraine, the appeal of an automated vulnerability hunter is straightforward. The more weaknesses defenders can identify and close before they are exploited, the less time attackers have to turn software flaws into disruptions of critical services.

Ukraine is not OpenAI’s first government cybersecurity partner. The company has already provided access to its models to cyber agencies in France, Germany and Poland, while CERT Polska has used the technology to identify six vulnerabilities in commercial router software. The flaws were subsequently patched by the vendor.

The Ukrainian deployment also puts OpenAI’s broader cybersecurity ambitions into practice. On September 3, the company pledged $1 billion in subsidized Daybreak access for resource-constrained cyber defenders around the world and said it would expand the program to partner countries within weeks.

Three weeks later, Ukraine became a visible example of what that promise could look like on the ground.

OpenAI’s Cybersecurity Role Is Moving Closer to the Battlefield

The Ukraine deployment also marks another step in OpenAI’s changing relationship with military and national-security work.

Until January 2024, OpenAI’s usage policy explicitly prohibited military and warfare applications. That language was removed from the policy page on Jan. 10, 2024, shortly before CEO Sam Altman and OpenAI’s vice president of global affairs said at the World Economic Forum in Davos that the company was already developing cybersecurity tools with the Pentagon.

Since then, the boundary between frontier AI and national security has continued to shift. Cybersecurity sits at a particularly important intersection: the technology can be used defensively to identify vulnerabilities, but the same underlying capabilities can also make it easier to analyze code and discover weaknesses.

That expansion has also raised questions about what technology companies gain from operating alongside governments confronting sophisticated cyber threats. Jamie MacColl, a cyber researcher at the Royal United Services Institute, told the BBC that Western technology companies can also gain valuable intelligence from an active war zone, noting that it was therefore unsurprising that OpenAI was providing defensive cybersecurity services.

OpenAI has not said that Ukraine is being used as a testing ground for its technology, and the deployment does not by itself establish such a motive. What is clear is that Daybreak is moving from an AI cybersecurity initiative into an environment where the stakes of finding a vulnerability first are unusually high.

For Kyiv, that race is ultimately about time. Every software weakness discovered and fixed before an attacker can exploit it represents one less opening into the systems that keep a country running.

And Ukraine is now putting OpenAI’s promise to work where that race is already underway.