What is OpenAI's Daybreak program in Ukraine?

OpenAI has extended its Daybreak cyber defense program to Ukraine's Ministry of Digital Transformation to help counter thousands of annual attacks on civilian infrastructure. This initiative provides access to advanced AI models like GPT 5.6 Sol for reviewing legacy software, investigating suspicious activity, and testing fixes before deployment.

How much does it cost for Ukraine to use OpenAI's Daybreak tools?

OpenAI has committed $1 billion in subsidized tokens to support its global initiative for critical infrastructure operators, including Ukraine. This funding effectively removes cost barriers for governments and agencies that might otherwise lack the budget to access cutting-edge AI models.

How many cyber attacks does Ukraine face annually?

In 2025 alone, CERT-UA handled nearly 6,000 cyber incidents, many of which targeted essential services like hospitals and energy grids. These persistent digital threats are part of broader Russian offensive cyber operations that have been a key component of aggression against Ukraine since 2014.

Does Ukraine share sensitive data with OpenAI through the Daybreak program?

The Daybreak program raises concerns about data intelligence gains for Western tech firms during an active conflict. Experts note that companies like OpenAI benefit from gathering real-world behavioral data on how attackers operate and how defenses respond under pressure.

How does the Daybreak program change OpenAI's role in national security?

This move marks a shift from commercial cloud use to active national security operations, integrating AI models into Ukraine’s national defense framework. It signals a broader strategy by OpenAI to deploy AI-powered defense tools at scale across critical infrastructure operators in nations under sustained cyber pressure.

OpenAI Daybreak in Ukraine: AI Cyber Defense Costs and Impact

TL;DR: OpenAI has deployed its Daybreak program to Ukraine’s Ministry of Digital Transformation, providing subsidized access to advanced models like GPT 5.6 Sol to counter the nearly 6,000 cyber incidents recorded in 2025. This strategic shift integrates AI directly into national defense infrastructure, aiming to protect critical civilian services from persistent Russian attacks while raising questions about data intelligence gains for Western tech firms.

Key facts

OpenAI extended its Daybreak cyber defense program to Ukraine’s Ministry of Digital Transformation, marking a shift from commercial cloud use to active national security operations.
The initiative provides access to advanced AI models, specifically the GPT 5.6 Sol model, for reviewing legacy software and investigating suspicious digital activity.
Ukraine’s CERT-UA handled nearly 6,000 cyber incidents in 2025, many targeting civilian infrastructure like hospitals and energy grids.
OpenAI committed $1 billion in subsidized tokens to support critical infrastructure operators globally, removing cost barriers for governments.
The program follows similar access grants to national cyber agencies in France, Germany, and Poland, signaling a broader strategy across EU nations.

OpenAI Expands Daybreak Program to Ukraine Amid Escalating Cyber Threats

In a significant shift from commercial cloud deployment to active national defense, OpenAI has extended its Daybreak cyber defense program to the Government of Ukraine. The initiative is being implemented in partnership with Ukraine’s Ministry of Digital Transformation [1][3]. This move comes as Ukrainian authorities face persistent digital threats targeting essential services like hospitals, energy grids, and telecommunications systems.

The extension follows similar access grants to national cyber agencies in France, Germany, and Poland, signaling a broader strategy by OpenAI to deploy AI-powered defense tools at scale across critical infrastructure operators [1]. The program now moves beyond AWS-based technical demonstrations into real-time support for a nation under sustained cyber pressure.

What the Tools Actually Do

Daybreak provides defenders with access to advanced AI models, including the GPT 5.6 Sol model, to perform three core functions: reviewing legacy software for weaknesses, investigating suspicious digital activity, and testing potential fixes before deployment [1][3]. These capabilities are designed to accelerate vulnerability detection and remediation—tasks that are often time-consuming and resource-intensive.

For example, Poland’s CERT Polska used the same tools to identify six vulnerabilities in third-party router software, which were later confirmed and patched by the vendor [1]. Similarly, ENISA leveraged Daybreak to resolve security flaws across multiple EU institutions [1]. These cases illustrate how AI can act as a force multiplier for under-resourced national cyber teams.

The Scale of the Threat: 6,000 Incidents in One Year

The urgency behind OpenAI’s expansion is underscored by data from CERT-UA, Ukraine’s national incident response team. In 2025 alone, the agency handled nearly 6,000 cyber incidents—many targeting civilian infrastructure [1][3]. These attacks are not isolated; Rafe Pilling of Sophos noted that Russian offensive cyber operations have been a key component of aggression against Ukraine since 2014 [3].

George Osborne, Head of OpenAI for Countries, said, according to the BBC: “Protecting civilian infrastructure means defending it against both physical and digital attacks, so people can continue to live, work and access essential services” [3]. The Daybreak program aims to close that gap by giving defenders faster tools to respond.

What It Costs: $1 Billion in Subsidized Tokens

OpenAI has committed $1 billion in subsidized tokens to support its global initiative for critical infrastructure operators, including Ukraine [2]. This funding effectively removes cost barriers for governments and agencies that might otherwise lack the budget to access cutting-edge AI models. The program is not a paid service but a strategic investment in global cyber resilience.

The term “defender’s window” describes OpenAI’s urgency framework: the belief that defensive AI tools must be deployed quickly—before malicious actors fully exploit similar capabilities [2]. This concept frames the current moment as a narrow opportunity to strengthen defenses at scale, especially in high-risk regions like Ukraine.

Unanswered Questions: Who Benefits from the Data?

While welcomed by Ukrainian officials and cybersecurity experts, the initiative raises concerns about data intelligence gains for Western tech firms. Jamie MacColl of RUSI suggested that companies like OpenAI benefit from gathering real-world behavioral data during an active conflict [3]. This includes how attackers operate, how defenses respond, and what vulnerabilities emerge under pressure—information that can refine future AI models.

Ukraine already uses AI tools provided by competitors such as Google, indicating a growing reliance on foreign technology for national defense [3]. The expansion of Daybreak into Ukraine’s civilian infrastructure operations marks a clear evolution—from supporting government workers via cloud platforms to enabling real-time cyber defense in wartime conditions.

A New Era of AI in National Security

This move represents a pivotal shift: OpenAI is no longer just offering AI tools for enterprise or research use. By integrating its models into Ukraine’s national defense framework, the company is effectively participating in active cyber defense operations. The extension of Daybreak to Ukraine is not merely an update—it is a strategic repositioning of AI from commercial infrastructure to geopolitical frontline defense.

As cyberattacks grow more frequent and sophisticated, OpenAI’s decision underscores a growing trend: AI is becoming a core component of national security strategy, especially in nations under sustained digital assault.

Sources

OpenAI extends cyber access to Ukraine for civilian defense | daily.dev — 2026-09-23
OpenAI gives AI cyber defence tools to Ukraine (www.bbc.com) — 2026-09-23
Event Replay: Daybreak: Strengthening Cyber Resilience with AI - Video | OpenAI Forum (forum.openai.com) — 2026-09-23