AI-first underwriting. For the cyber line that finally needs it.

External scanning tells you what an attacker sees. Draxis is an AI-first platform that reads what each policyholder’s controls are doing, continuously and not at renewal, and projects the patterns up across your whole book.

Request a briefing →
The underwriting problem

You are pricing risk from a point-in-time snapshot of a changing environment.

Renewal questionnaires reflect intent, not reality. External scan scores reflect what is visible to the internet, not what is happening inside the control environment.

The gap between what carriers know at underwriting and what is actually true six months later is where claims come from.

Questionnaire dependency

Underwriting still relies primarily on self-reported security posture data. Policyholders answer optimistically. Controls degrade between renewals. The data ages out the moment it is submitted.

Outside-in blind spots

External scan vendors surface what is visible from the internet. MFA enforcement gaps, EDR coverage failures, over-privileged identities, and lateral movement paths are invisible to outside-in tooling. These are the controls that determine claim outcomes.

Point-in-time underwriting

Risk posture is dynamic. A policyholder who passes renewal screening in January may have material control gaps by March. Carriers have no signal between checkpoints. The first indication of degradation is the claim.

How Draxis is different

Inside-out risk signals. Not another outside-in score.

Capability External scanning
Bitsight, SecurityScorecard, RiskRecon
Draxis
Signal source Public internet surface Internal control telemetry
MFA enforcement gaps Not visible Detected and quantified
EDR coverage by device Not visible Detected and quantified
Identity hygiene Not visible Detected and quantified
Conditional access gaps Not visible Detected and quantified
Open ports / exposed services Visible Not primary focus
TLS / certificate health Visible Not primary focus
Continuous monitoring Score updated periodically KRIs updated per control sync
Financial exposure mapping Not provided Mapped per finding, per tenant
Framework alignment Not provided NIST CSF, SOC 2, ISO 27001, SEC Cyber Rules, DORA, NIS2
Questionnaire replacement No Partial. Structured KRI evidence replaces subjective self-reporting for covered control domains

Draxis and external scanning vendors are complementary, not competing. External scanning covers the attack surface. Draxis covers the control environment. Both signals improve underwriting precision when used together.

Use cases

How carriers and brokers put AI to work today.

AI-drafted underwriting narratives

For each insured, AI assembles an underwriting-grade narrative: current control posture, scenario-level expected losses tied to coverage triggers, drift since binding, and recommended renewal terms. Verifiable evidence, not self-attestation.

AI portfolio early-warning

When AI detects a pattern in one insured that historically preceded claims in similar insureds, it alerts you with the cohort context. Your book itself becomes an early-warning system, long before the next renewal cycle.

AI loss-event ↔ policy mapping

AI maps catalog loss events to your policy language (sublimits, retentions, exclusions) so scenario outputs translate directly into expected indemnity, not abstract dollars. Cross-portfolio shifts summarized for underwriters in plain English.

AI-powered broker differentiation

Broker advisory practices use Draxis as the AI engine behind their SMB cyber risk service. An AI vCISO and expert panel under your brand, producing the renewal narrative your insureds’ carriers actually want to read.

Deployment model

Draxis fits into your existing carrier infrastructure.

Step 1 · Policyholder consent

Policyholder authorizes read-only access

At policy inception or renewal, the policyholder grants Draxis read-only API access to one or more of their existing security controls. Authorization takes under 10 minutes. Nothing is installed in the policyholder’s environment.

Step 2 · Continuous KRI extraction

Draxis extracts and monitors KRIs

Draxis reads from the connected controls on a configurable cadence: daily, weekly, or per-event. Findings are mapped to financial exposure and framework alignment automatically. Material changes trigger alerts.

Step 3 · Carrier portal integration

Signals surface in your workflow

KRI data is available via API to your existing policyholder portal, underwriting workflow, or data warehouse. Draxis does not require carriers to replace existing tooling. It adds the inside-out signal layer to whatever infrastructure you already operate.

Pilot structure: Draxis works with carriers and MGAs on bounded pilots, typically 25–100 policyholders over 90 days, with defined success metrics tied to underwriting outcomes or loss ratio signal. Contact us to discuss a pilot framework.

This is a conversation, not a demo request.

If you are evaluating risk intelligence capabilities for your cyber book, or building a policyholder portal and looking for a continuous signal layer, we would like to speak with you.

We do not do automated demos for this segment. We do structured briefings with the right people on both sides.

Request a briefing →

Or reach us directly at carriers@draxis.ai