DeRISK UWA
An OT submission takes 10–30 hours to assess by hand. DeRISK UWA turns it into an underwriting verdict, actuarial loss outputs and binding conditions in 10–20 minutes — with a full audit trail and an underwriter reviewing every call.
Submission · Power generation, 4 sites
Underwriting verdict
Quote with conditions
2 binding conditions · underwriter review required
Occurrence exceedance (OEP)
Actuarial engine · 100,000 Monte Carlo iterations
Model confidence
Request supplemental OT data before binding
Submission to verdict
18 min 10–30 h by hand
10–20
minutes to actuarial output
100,000
Monte Carlo iterations per run
9
pre-trained questionnaire formats, no field mapping
31+
threat intelligence collectors
The questionnaire and every document attached to it are evaluated together, so findings reflect the complete submission. Work that takes 10–30 hours by hand comes back in minutes.
DeRISK UWA reads 9 pre-trained insurer and broker questionnaire formats, plus OT/ICS and ransomware supplementals, with no field mapping. Every document passes four extraction steps before it is assessed.
A submission that cannot evidence its OT controls is a risk signal, not a neutral data gap. Stated and verified controls are marked apart, and IT and OT each carry their own confidence score.
A 100,000-iteration Monte Carlo returns expected annual loss (EAL), value at risk (VaR), occurrence and aggregate exceedance probabilities (OEP/AEP) and maximum foreseeable loss (MFL). Enterprise tenants can substitute their own actuarial model entirely: your pricing model, our risk intelligence.
Every assessment ends in a verdict with a premium indication, binding conditions with deadlines and an audit trail behind each figure. A mandatory human review step sits before anything is bound.
Each agent does one job and hands the result to the next. Nothing is re-keyed from the questionnaire, and every figure in the verdict traces back to the agent that produced it.
See DeRISK UWA in action
How a submission becomes an underwriting verdict in 10–20 minutes.
Every submission, in 10–20 minutes. Approve, quote with conditions or decline, with binding conditions, deadlines and separate IT and OT confidence.
See the verdict →Every submission. EAL, VaR, OEP/AEP and MFL from a 100,000-iteration Monte Carlo, with the methodology behind each figure.
See the loss output →Every submission. A premium indication and a structured insurance program, priced from the same loss model the actuary reviews.
See the program structure →Every report run is recorded. Each control shows whether it was evidenced or only attested, and every figure traces to its source document.
Three tiers for what you submit and what comes back, and five premium capabilities that attach to them.
Freemium
A real assessment on one submission, at no cost.
Premium
Assessments that stand up at an underwriting committee.
Premium
Our risk intelligence inside your own pricing model.
Attach to Professional or Enterprise. Click an add-on to open it.
Ask DeRISK. Davy answers.
Ask about any submission in plain words.
Works across
and DeRISK Threat Intelligence
Remediation for the insured, in order.
Remediation the insured can act on, in order.
Expert judgment, encoded across
Test a condition before you bind.
Test a condition before you bind.
Four scenario dimensions
Premium OT feeds, bundled by DeNexus.
Premium OT threat feeds, bundled by DeNexus.
The feeds you already pay for.
Your own threat-intel subscriptions, in every assessment.
Each premium tier includes everything in the tier below it. What changes is how much you can submit, what the report carries and whose actuarial model prices it.
Each capability is priced separately and attaches to Professional or Enterprise. None attach to Standard.
What underwriters ask before the first submission — the output, the evidence behind it and what it costs.
Talk to salesA completed questionnaire is enough to start. DeRISK UWA is pre-trained on the standard broker and carrier submission formats, including OT and ransomware supplementals, so there is no field mapping to do. Proprietary formats are trained on request.
On premium tiers you can attach complementary documents alongside the questionnaire, such as asset inventories, security policies and prior assessment reports. The evaluation waits until every document is processed and then reads them together, so the findings reflect the whole submission rather than the questionnaire alone.
Between 10 and 20 minutes from submission to finished assessment, against 10 to 30 hours of manual underwriting work. The run executes server-side, so closing the browser tab or losing your connection does not interrupt it.
An underwriting assessment with a verdict — approve, quote with conditions or decline — backed by expected loss, maximum foreseeable loss and occurrence exceedance curves. It includes a premium indication, a structured insurance programme and binding conditions with deadlines.
The report runs from the verdict through the risk profile, the quantification and the recommendation to the evidence behind them, and closes with a full audit trail and attribution statement.
It counts against the risk. A submission that cannot evidence its OT controls is treated as a risk signal, not a neutral gap — attested controls carry less weight than evidenced ones.
Every assessment carries separate confidence scores for IT and OT, marks which controls were only stated in the questionnaire and which were verified, and recommends collecting supplemental data before binding when confidence is low.
Yes. Human review is a mandatory step in every assessment, and every figure carries its audit trail, so the underwriter can see where each number came from before accepting it. DeRISK UWA is built for NAIC, FCA, Lloyd's and EIOPA expectations for AI in underwriting.
Yes, on Enterprise. By default, assessments run on the DeNexus actuarial engine, 100,000 Monte Carlo iterations with no setup required. Enterprise lets you substitute your own actuarial model entirely: your pricing model, our risk intelligence.
Standard is free: single-document submissions, the pre-trained questionnaires, standard threat intelligence, NIST CSF and IEC 62443-inspired frameworks, the standard actuarial engine and a short report.
Professional adds multiple-document submissions, custom submission types, committee-ready and white-labeled reports, report persistence and your own Insurer Specific Framework. Enterprise adds API access, single-tenant deployment, your own actuarial model and your customized report. Pricing is quoted rather than published — talk to sales for a figure against your submission volume.
Separately, and à la carte. Five capabilities attach to Professional or Enterprise — DeRISK Davy, the DeRISK Consultant, the DeRISK Scenario Simulator, the Threat-Intel Premium Feed and Threat-Intel: Bring Your Own. You take the ones you need and none of them is bundled into the tier price.
None attach to Standard. And the two CTI add-ons extend the threat intelligence that ships with every tier, including Standard — they never replace it.
Your team can sign in with their Microsoft work accounts through Entra ID, with email and password still available. The DeRISK UWA app for Microsoft Teams lets underwriters submit questionnaires, track processing and download reports without leaving Teams. Enterprise adds API access for connecting DeRISK UWA to your own systems.
No. The agents that assess your submissions are stateless consumers of your data — nothing you supply becomes part of any model, so there is nothing learned from it to remove.
Inference runs privately inside DeNexus’ own cloud environment: outputs are not used to train or improve any model. Invocation logging is disabled in our environment. Your submissions persist only in DeNexus's own governed storage, where standard deletion controls apply.
Start free on DeRISK™ UWA Standard with a VRI short report on a single submission. Move to Professional for committee-ready reports, or to Enterprise for API access and your own actuarial model.