The insurance industry has been assessing technology risk with questionnaires designed for an era when the biggest question was "do you have a firewall?" That approach was already showing its age for cyber risk. For AI risk, it's completely inadequate.
We built VectorIQ™ because the industry needs a structured, repeatable, data-driven way to evaluate AI risk — one that produces scores carriers can actually underwrite against, not just narratives that get filed away.
The Problem with Generic Questionnaires
- ❌ Generic yes/no questions that don't capture nuance
- ❌ No industry-specific weighting
- ❌ No scoring — just a narrative summary
- ❌ Doesn't distinguish between low-risk and high-risk AI usage
- ❌ Can't map to specific coverage lines
- ❌ Takes weeks to complete and interpret
- ✅ Graduated 0–5 scoring with specific anchors per question
- ✅ Dynamic industry-specific weight adjustment
- ✅ Produces a numerical score carriers can underwrite against
- ✅ Differentiates AI use cases by harm potential
- ✅ Maps risk dimensions to specific coverage lines
- ✅ 4-minute assessment, instant scoring
The Four Modules of VectorIQ™
VectorIQ™ evaluates AI risk through four interconnected modules, each designed to capture a different dimension of exposure. The modules interact — a high harm score with strong controls produces a very different risk profile than a moderate harm score with no controls at all.
Harm Potential
Evaluates what could go wrong. Covers seven dimensions including physical safety, discrimination and bias, privacy violations, financial harm, IP/content risk, regulatory exposure, and systemic/correlated risk. Each dimension is weighted dynamically based on industry.
Control Effectiveness
Measures what safeguards are in place. Evaluates human oversight, testing and monitoring, data governance, vendor management, documentation, incident response, and employee training. Strong controls meaningfully reduce the net risk score.
Exposure Amplifiers
Quantifies factors that scale the risk. Includes volume of AI decisions, autonomy level, data sensitivity, regulatory jurisdiction, and third-party dependencies. These amplifiers don't create risk on their own — they magnify whatever harm potential exists.
Adjuster Load
Accounts for claims complexity. Evaluates explainability of AI decisions, audit trail quality, regulatory defensibility, and litigation complexity. Higher adjuster load means claims will be more expensive and harder to resolve — even if the underlying risk is moderate.
How the Score Is Calculated
The VectorIQ™ score isn't a simple average. It's a formula that reflects how these risk dimensions actually interact in real-world claims scenarios:
This formula captures several important realities. Controls reduce risk but can't eliminate it — a company using AI to make autonomous medical decisions has inherent harm potential that no amount of governance fully offsets. Exposure amplifiers compound harm multiplicatively — processing 1 million AI decisions a month is fundamentally different from processing 1,000. And adjuster load adds to the total independent of the other factors — even a moderate-risk scenario becomes expensive when AI decisions are unexplainable.
Industry-Dynamic Weighting
A law firm's AI risk profile looks nothing like a fintech company's. VectorIQ™ adjusts its weighting tables based on industry, so the same AI chatbot scores differently depending on whether it's answering general customer questions for a retailer or providing guidance to patients for a healthcare company.
Example: In healthcare, the physical safety and regulatory dimensions are weighted heavily. In HR and staffing, bias and discrimination dimensions carry more weight. In fintech, financial harm and regulatory compliance dominate. The questionnaire is the same — but the scoring reflects the reality that risk materializes differently in every industry.
What Makes This Patent-Pending
VectorIQ™'s patent-pending methodology covers several novel elements: the multi-dimensional scoring architecture that maps AI risk dimensions to specific insurance coverage lines; the dynamic industry-weighting system that adjusts risk evaluation in real-time based on sector-specific exposure patterns; the formula architecture that models how harm potential, controls, exposure, and adjuster complexity interact to produce a single underwriting-grade score; and the automated modifier system that derives risk surcharges and discounts directly from intake responses without manual underwriter intervention.
This isn't just a better questionnaire. It's a new framework for understanding and quantifying AI risk in commercial insurance — one that gives carriers, brokers, and businesses a common language for a risk category that didn't exist five years ago.
See VectorIQ™ in action.
Take the 4-minute assessment and get your score — with a full breakdown of where your AI exposure sits across all four modules.
Start Your Assessment →