OpenAI and ReliaQuest: Partnering for Agentic Cybersecurity
US cybersecurity company ReliaQuest has partnered with OpenAI to accelerate the development and deployment of advanced AI capabilities for enterprise cyber defence.
Under the partnership, ReliaQuest will receive access to advanced OpenAI models and cyber capabilities to support rapid development within its platform.
AI is increasingly used on both the defensive and adversarial side of cybersecurity. Anthropic recently published evidence showing that attackers' use of AI is increasing the threat they pose to other organisations.
“AI is already good and about to get super good at cybersecurity; we'd like to start working with as many companies as possible now to help them continuously secure themselves. ”
OpenAI partnership
Working together, OpenAI and ReliaQuest will jointly identify and prioritise high-impact security operations use cases, then collaborate through development, testing and evaluation and production rollout within ReliaQuest’s GreyMatter platform and associated managed workflows.
ReliaQuest says it will contribute expertise and real-world experience to help codify standards for safety and abuse prevention, as well as the right controls to monitor and prevent unsanctioned activity.
Brian Murphy, Founder and CEO of ReliaQuest, says: “Cybersecurity is being reshaped by AI on both sides.
“The more we can collaborate, the better we can defend against AI-driven attacks. This partnership with OpenAI helps us move faster in bringing the most advanced Agentic AI capabilities into GreyMatter, so security teams can investigate and respond to threats in minutes.”
OpenAI’s Daybreak
To partner with the US AI giant, ReliaQuest joined its OpenAI Daybreak Cyber Partner Program. According to information on OpenAI’s website, some of its other partners include Palantir, Cisco, Accenture, Crowdstrike, Darktrace, Sophos and SoftBank.
For the programme, OpenAI says it is working with the world’s leading cyber organisations as partners to bring defensive capability into the tools, services and workflows that security teams already rely on.
Announcing the programme on X in May of 2026, OpenAI’s CEO, Sam Altman said: “OpenAI is launching Daybreak, our effort to accelerate cyber defence and continuously secure software.
“AI is already good and about to get super good at cybersecurity; we'd like to start working with as many companies as possible now to help them continuously secure themselves.”
OpenAI’s cyber capabilities
The partner program will provide ReliaQuest with closer technical collaboration and access to OpenAI’s frontier cyber capabilities and models.
OpenAI’s CyberSecurity capabilities include the CodexSecurity plugin and GPT‑5.5‑Cyber, which can be used for authorised red teaming (simulating real world attacks), penetration testing, exploit validation and controlled security testing.
ReliaQuest says the access can help enterprises find and patch vulnerabilities faster, while enabling security teams to keep pace as adversaries increasingly use AI to automate and scale attacks.
- Anthropic said that malicious actors are using AI in ways that make them more dangerous
- OpenAI’s CyberSecurity capabilities include the CodexSecurity plugin and GPT‑5.5‑Cyber
- 67% of accounts studied by Anthropic used AI when preparing for a cyberattack
Growing cyber threats from AI
In June, another leading US AI firm, Anthropic, shared a blog post analysing how well the techniques and frameworks used by the security community are holding up in an era dominated by AI.
In the post titled ‘What we learned mapping a year’s worth of AI-enabled cyber threats,’ Anthropic shared some key conclusions.
Anthropic said that malicious actors are using AI in ways that make them more dangerous. More specifically, it said threat actors are using AI in the later, more complex stages of their cyber operations.
The company said cyberattacks are becoming more autonomous and the fact that AI can be used to chain together many parts of the attack means that the old ways of differentiating high from low-risk actors are no longer as effective.
Anthropic noted that the most common AI-enabled activities in its database related to preparing for a cyberattack, such as writing malware. 560 of the 832 accounts it studied, or 67%, used AI for this purpose.
Its evidence suggested AI was increasing the threat level from attackers. It said 33% of actors were classified by its risk-scoring system as medium risk or higher. But by the second six-month period, that share had jumped to 56%.
