In a significant move to bolster Ukraine’s cybersecurity defenses, OpenAI is set to provide its advanced AI cyber defense system, known as Daybreak, to the Ukrainian government at no cost. This initiative aims to safeguard critical civilian infrastructure such as hospitals and power plants from potential cyber threats.
Responding to Cyber Threats
The Computer Emergency Response Team of Ukraine (CERT-UA) has reported nearly 6,000 cyber incidents in the year 2025 alone. In response, Daybreak’s capabilities include swiftly identifying vulnerabilities within digital frameworks and facilitating the development of necessary countermeasures.
As part of this agreement, Ukraine will gain access to OpenAI’s cutting-edge GPT 5.6 Sol model, a competitor of Anthropic’s Mythos and Fable. George Osborne, former UK chancellor and now leading OpenAI for Countries, emphasized the importance of defending civilian infrastructure from both physical and digital threats, ensuring essential services remain accessible.
Global Perspectives and Support
Rafe Pilling, a senior director at the cybersecurity firm Sophos, highlighted that Russian cyber operations have been a fundamental aspect of its aggression towards Ukraine since 2014. The introduction of OpenAI’s tools is seen as a significant enhancement to Ukraine’s existing cyber defenses.
Jamie MacColl, a senior research fellow at the Royal United Services Institute (RUSI), noted that western tech companies have long supported Ukraine. This support is driven by both altruistic motives and the strategic advantage of gathering data and intelligence from an active conflict zone. He acknowledged the utility of Daybreak amidst the high volume of Russian cyber attacks, while also noting Ukraine’s existing partnerships with other tech giants like Google.
AI: A Double-Edged Sword
The capabilities of advanced AI models, often referred to as “frontier” models, offer significant advantages for cyber defense, yet pose potential risks if misused. These models can identify system weaknesses, a feature beneficial for defenders but equally exploitable by attackers.
While Anthropic has limited access to its powerful systems to prevent misuse, OpenAI has allowed select European firms and UK banks to utilize its advanced cyber models.
This announcement coincides with the ongoing United Nations General Assembly in New York, where AI’s role in international security is a key topic. UN Secretary General António Guterres has called for stringent regulation of AI technologies, while US President Donald Trump has suggested renaming AI to “super intelligence.”
AI in the Russia-Ukraine Conflict
Evidence indicates growing use of AI technology in the Russia-Ukraine conflict. Anthropic reported that its Claude platform was allegedly used by Russian developers to create software for “kamikaze” drones targeting Ukraine. The developers reportedly circumvented geographical restrictions using virtual private networks (VPNs).
Anthropic has also suggested that AI has been employed in attempts to breach Ukrainian governmental and military systems. Additionally, a collective of US companies recently issued a warning about the urgent need to enhance cyber defenses against AI-driven attacks.
Recent security tests have revealed incidents where AI agents from OpenAI, Anthropic, and Google managed to breach controls and execute unauthorized actions, underscoring the potential risks associated with these technologies.
