Brinqa Joins Anthropic's Cyber Verification Program
by Dan Pagel, CEO & Board Director//6 min read/
Update: Since this post was published, Brinqa has also been accepted into OpenAI's Trusted Access for Cyber program.
Better data makes better AI, and better AI makes better outcomes, faster. We built the foundation that makes that chain real.
Every security vendor is about to tell you they have frontier AI. Most of them mean they have access to a model. Access is not the hard part, and it is about to be everywhere. What these models are genuinely good at is finding more vulnerabilities, more connections between them, and more exposure than any team could previously see. That is not a small thing, but it makes the data problem harder, not easier. More intelligence without the foundation to act on it is just faster noise. The model is only as useful as what you point it at. The harder problem is what to do with what it finds.
Access Is Not the Hard Part
Brinqa has been accepted into Anthropic's Cyber Verification Program, which gives vetted organizations access to Anthropic's most advanced capabilities for legitimate defensive work, under Anthropic's security and responsible-use controls. We're glad to be in it and we take the partnership seriously. What we won't do is oversell it.
You are going to see vendors turn program access into headlines that suggest Anthropic hand-picked them for a mission. It didn't. The program grants access to advanced capabilities for defensive work. That is worth having. It is not a coronation, and dressing it up as one just adds noise to a market that already drowns buyers in it. The same way a model hands you a confident answer built on bad data, a vendor hands you a confident headline built on a press release. We would rather tell you what this actually is and let the work carry the rest.
"The race everyone can see is for access to frontier AI. The race that actually decides who wins is for the data foundation that makes it actionable. We ran that one first," said Brad Hibbert, Chief Strategy Officer at Brinqa.
The Data Problem AI Makes Worse
That foundation is what separates a vulnerability management program from a mature exposure management program. A model gives you an answer at the same confidence whether the data behind it is complete or garbage. Ask it whether a vulnerability is exploitable and it will answer with total conviction, even if it has never seen your asset relationships, your business context, or your existing controls. It is a doctor diagnosing you without the lab results, certain of itself and possibly wrong. Better AI starts with better data, or it just produces faster mistakes.
That is why we built the data foundation first, years before a program like this existed. The Brinqa CyberRisk Graph correlates hundreds of millions of findings across your assets, enriched with business context and threat intelligence, and mapped to the relationships that actually exist in your environment. BrinqaDL holds it as a curated system of record. This is the part a new entrant cannot stand up no matter how much it raises, because depth like this is earned across years of real environments, not bought. Our clients are seeing 50 to 200 percent more findings than they were a year ago, and the foundation is what turns that volume into something a team can act on instead of drowning in.
On top of that foundation, our agents already do the unglamorous work that makes the data trustworthy and actionable, assigning the right remediation owner and clearing duplicate findings, so the AI reasons over clean signal instead of noise.
The model itself is where we made a different bet than most. We don't think the durable value is in owning the model. It's in owning the layer the model has to run on. So through our BYOAI architecture, you can bring your own AI, including your Anthropic agents, and point it at the Brinqa data layer over MCP. Your environment, your relationships, your remediation options, at scale. Closed boxes age the moment a better model ships. An open layer gets better with every model that comes after it. That is not a feature, it is a posture, and it is the difference between betting on one model and betting on all of them.
None of this is a bolt-on. The platform was built to move from orchestrating data, to orchestrating decisions, to orchestrating action, because remediation isn't a security outcome alone. It takes security and IT operations agreeing on the same risk picture. Brinqa’s AI Exploitability Agent brings adaptive prioritization with calculated reachability, and our AI Remediation Agent drives the remediation campaigns that close the loop, both landing this quarter. We are not describing a roadmap we hope to build. We are describing the system the foundation was always pointed at.
The Race That Actually Decides Who Wins
Cheap, capable models are coming fast, and many will arrive without guardrails. The teams that come out ahead won't be the ones who got access first. They will be the ones who built the foundation that makes frontier AI dependable instead of confidently wrong. Better data, better AI, better outcomes, faster. That is the chain. We built the start of it years ago. The Cyber Verification Program is how we keep sharpening the end.
Your exposure management program is only as strong as the data underneath it. If you want to see what that looks like in practice, meet with our team.
