AI system inventory
A per-insured list of the AI systems found and what each one can do.
Exposure Discovery
Find every AI system already sitting in your book.
Policies written before AI existed now cover it silently. See which insureds run AI systems, what those systems can do, which policies respond, and where the exposure concentrates.
Built for real underwriting work.
Every insured on the book is checked against public signals to find the AI systems they run, declared or not. Results break down by line, by insured, by policy and by individual system.
For each system, what it can actually do: what it reads, what it changes, whether it can move money, and whether any action is irreversible. Not only AI agents, but any AI system the insured has put in front of customers or decisions.
Those capabilities are matched to the policies already on the account, so you see which clause responds if the system causes the loss, and where the wording is silent.
Once systems are assessed, the ones that share a model, a vendor or a channel are flagged, because they fail together.
Everything around it.
The settings, outputs and controls that come with it.
Declared vs. found
Side by side: what the application said, and what is running.
Line exposure score
One number per line, comparable across the book.
Vendor concentration
Which platforms and models sit behind the most AI agents.
Change alerts
A note when an insured adds, removes or swaps an AI agent.
Policy response
Which clause on the account responds, and where the wording is silent.
Filters
Slice by line, region, premium band or AI agent type.
Evidence links
Every finding links to the signal it came from.