Hypernative has been accepted into Anthropic's Cyber Verification Program. Verification like this is not a formality, and it is not broadly given. Anthropic reviews each applicant individually, and approval is limited to organizations that can show a legitimate, ongoing need to reason about dual-use security material, the kind of work most general-purpose models are built to restrict by default.
Anthropic built the program around a narrow problem. The reasoning that lets a defender understand an attack first is the same reasoning an attacker would use to run it, so models default to blocking or limiting analysis of dual-use material, such as vulnerability exploitation research, regardless of who is asking or why.
Programs like this matter because they let verified defenders reason at the depth this work actually requires. That depth only means something if you already have the detection models and the track record to put it to use, and few companies in onchain security can bring both to the table. For us, verification means combining years of onchain security expertise with frontier model reasoning, and doing even more for the protocols and institutions we protect.
Gal Sagie, co-founder and CEO @ Hypernative
What this unlocks for Hypernative
Verification changes what's possible in three areas:
- Real-time exploit triage. Hypernative's proprietary detection models flag the suspected attack. From there, Claude can confirm whether it is an actual exploit and reconstruct what happened, root cause, attacker path, impact, into an analyst-grade incident writeup. That verification and writeup work requires reading attacker bytecode and modeling the pattern live, exactly the kind of task the safeguards used to limit or refuse outright. Verification lifts that constraint.
- Detection research. This involves reasoning over past exploits and attacker bytecode to model new attack patterns and enhance what Hypernative's proprietary models and detection agents catch going forward.
- Threat intelligence. It helps with deeper root-cause analysis behind Hypernative's Incident Reports and Developing Stories, enhancing analysis at all levels, from a single signal to the attacker's broader pattern.
What this means for our customers
For teams protecting onchain protocols, assets, or treasuries, this sharpens the intelligence they rely on when it matters most, during a live attack. When an exploit targets a protocol they are responsible for, Hypernative will be able to provide context even faster, offering more detailed root cause explanations and much more.
Reach out for a demo of Hypernative’s solutions, tune into Hypernative’s blog and our social channels to keep up with the latest on cybersecurity in Web3.
Proactive security for onchain finance.
