Cyber, privacy, & AI risk intelligence. one platform.
Score every submission for cyber, privacy and AI exposure without a questionnaire, see what’s exploitable across your book, and give policyholders a loss-control program backed by real people.

Trusted by leading carriers and brokers
I just wanted to share a sincere thank you for the work you and the team did this week on the PHP vulnerability. We really appreciate the support and partnership here.
Chief Operating Officer · White glove support


See what your desk gets.
One platform for cyber, privacy and AI risk, built for carriers, MGAs, reinsurers and brokers, and the policyholders they protect. From the first submission to the whole book.
Underwrite the exposure, not the questionnaire.
A Security Score, a Critical & High Risk Summary and every finding behind it, generated fresh at scan time. No agents, credentials or applicant cooperation.
Explore Cyber Risk Insights → Get the sample report →Score
0–1000
Underwrite what the website actually does.
Trackers, consent failures and data sharing, including what keeps firing after Reject All. Read it side by side with the cyber scan.
Explore Privacy Risk Insights → Get the sample report →Underwrite the AI, not just the company.
See how an applicant uses AI and what it means for each coverage line. The report checks the application against public evidence, lists the questions to ask before bind, and recommends Write, Write with conditions or Refer. Every finding links to its source.
Explore AI Risk Insights → Get the sample report →From the whole book to each insured.
RED tests known exploited vulnerabilities across every policyholder, and ZED flags exploited zero-days before they reach CISA’s catalog. CyRisk Protect then helps insureds close the gaps.
Explore Portfolio Management → Explore Loss Control →From domain to decision.
A domain is all we need.
Outside-in cyber, privacy and AI analysis, fresh at scan time.
Scores, sourced findings and a recommended disposition.
Monitor the book with RED and ZED, and close gaps with CyRisk Protect and expert help.
A team that picks up the phone.
Walkthroughs of individual scan observations, calls with your policyholders, and guidance from our staff privacy attorney.
See White Glove Support →Experts for the gaps.
Advisory and hands-on services, on their own or alongside CyRisk Protect.
Answers for cyber underwriters.
What outside-in data can and can’t replace on a cyber application.
Underwrite without a questionnaire →Check every insured against CISA’s Known Exploited Vulnerabilities catalog.
CISA KEV book check →Frequently asked questions.
What is CyRisk?
CyRisk is a cyber, privacy and AI risk intelligence company built for the insurance industry. Carriers, MGAs, reinsurers and brokers use it to underwrite, monitor portfolios and reduce policyholder risk.
Do we need a questionnaire to underwrite with CyRisk?
No. CyRisk reads each submission’s internet-facing security and privacy posture from the outside, with no agents, credentials or applicant cooperation. Many underwriters keep a short application for internal controls a scan can’t see.
What does the 0–1000 score mean?
It is one consistent scale you can tier and price against, where higher is safer. Category dials estimate how likely each type of loss is, and the overall score shows how exposed the weakest asset is.
How is CyRisk different from security ratings vendors?
CyRisk is built only for insurance. It reads cyber, privacy and AI exposure at the point of underwriting, finds confirmed exploitable exposure across the book with RED, catches pre-catalog zero-days with ZED, and gives policyholders a loss-control program backed by expert services.
Can CyRisk assess AI risk?
Yes. AI Risk Insights shows how an applicant uses AI, what it means for each coverage line, and whether to write, write with conditions or refer, with every finding linked to its source.
How can we try CyRisk?
Request a sample report, back-test five past submissions against the calls you made, or book a demo with our team.
Back-test us against your own book.
Send five past submissions you've already bound, conditioned or declined. We score them blind and put our read next to the call you made.









