Claims intake and coverage look-ups, without the typing
OpenAdapt learns claim entry from a demonstration, then enters each claim as a program: the same steps every time. By default each entry waits for approval, and it stops when something doesn't match. On openIMIS, an open-source insurance system, it saved 5 of 6 synthetic claims and stopped before saving the one with a mistyped insured number.
Book a 30-minute callSee the processes
Updated 2026-10-18
What the claims film shows
On 12 Oct 2026, a program learned from a demonstration entered 6 synthetic health-facility claims into a test copy of openIMIS as one worklist run, one claim after another, with no AI model calls.
- 5 of 6claims saved; our test setup read each back, 15 of 15 fields matching
- 1 of 6stopped before saving: a mistyped insured number
- 0AI model calls
On the 6th claim, the insured number had two digits swapped. openIMIS found no member, the screen no longer matched the demonstration, and the program stopped before saving. No claim was written.
Two things to know. The read-back was done by our test setup; this program has no record check of its own yet, so the engine labels the saved claims "Finished, not checked". On timing, a test script typing at a person's pace took 42.3 seconds to the Save click, and OpenAdapt took 47 to 73 seconds a claim on a shared test machine running openIMIS under emulation.
How long does your team take per claim today? We found no published study of insurer staff keying claims, so we time your own work in the access check.
Who it's for
Carriers, MGAs and TPAs whose staff key claims, first notices of loss or coverage look-ups into their own claims or policy system, where a read-only database role, a report or an API can confirm the save.
Brokers and agencies that key claims or submissions into carrier portals. We haven't tested a carrier portal yet. There, the check would be to open the saved claim again in a fresh browser session and compare every field with what was entered, set up with the workflow. How to stop re-keying claims covers the other options.
If claim documents arrive as PDFs, images or email, OpenAdapt can read them and fill in the entry's details, holding back any unclear field for a person to confirm. Document reading has been tested on synthetic documents only.
Processes
- Tested
Claims intake
openIMIS, 6 synthetic claims: 5 saved and read back, 1 stopped before saving.
Claims intake - Tested
Eligibility and coverage look-ups
openIMIS: 3 of 3 active policies confirmed, 3 of 3 expired policies stopped.
Coverage look-ups - Not recorded yet
Policy changes
Endorsements and cancellations keyed into a policy system. Not recorded on any system yet.
- Not recorded yet
Carrier-portal re-keying
One submission keyed into several carrier portals, each save to be checked by opening it again in a fresh browser session. Not recorded on any portal yet.
- Not recorded yet
Commission reconciliation
Carrier statements matched to the agency system. Not recorded on any system yet.
Can it check the record in your setup?
| Your setup | Second read | Result OpenAdapt can give |
|---|---|---|
| Carrier, MGA or TPA entering into its own claims or policy system | A read-only database role, a report export or an API | Done and checked, at the time of the run |
| Broker entering into its own agency system | The agency system's API or a report | Done and checked |
| Broker entering into a carrier portal | The saved claim, opened again in a fresh browser session | Checked on screen, at the time of the run (not yet tested on a carrier portal) |
| A look-up in a third party's portal | A second source, such as an eligibility file | Done and checked only with that source; otherwise a person reviews |
| The only read is the same screen | None | It stops, or ends "Finished, not checked" |
Questions insurers ask
Is AI making decisions about our members?
No. In our recorded runs the program made 0 AI model calls, and by default each save waits for approval. That matters under the NAIC model bulletin on AI use, adopted in 29 jurisdictions, and OSFI E-23 in Canada from May 2027.
How does it fit NAIC Model #668 and NYDFS Part 500?
Normal runs happen on your computer, and OpenAdapt Cloud receives results and receipts without member details. We don't hold a SOC 2 report; we answer your vendor questionnaire directly. See Security.
We're federally regulated in Canada. What about OSFI B-10?
The engine is MIT-licensed and runs locally, and you keep the programs and run records if you stop. We answer B-10 outsourcing questions, including the exit plan, in your review.
We're a health plan or TPA. Will you sign a BAA?
Yes, before any work on real data.
Is a program using our login allowed by portal terms?
Some carrier-portal terms forbid letting anyone else use your login. We read each portal's terms with you and prefer carrier APIs and downloads.
Who pays if it enters the wrong thing?
It checks against the record where it can, stops when it can't tell, and by default each save waits for approval. Liability terms are set in your pilot agreement; talk to your E&O carrier too.
Ready for your system, and what it's been tested on
Built for production: signed admissions, qualification against your normal and fault cases, a record check on every save, local or in your VPC, and a BAA on Enterprise. We qualify it on your system before it touches a real record.
Tested on openIMIS, an open-source insurance system, with synthetic claims. Not yet run on a commercial claims or policy system or a carrier portal.
Free to run locally. Team $499 a month. Healthcare and enterprise qualification sprints from $15k. Pricing.
Sources
- OpenAdapt on openIMIS 25.10 (open-source insurance system), local test copy, 12 Oct 2026. A program learned from one demonstration entered 6 synthetic health-facility claims in one batch, with 0 AI model calls. 5 were saved; our test setup read each back through SQL and the openIMIS API, and 15 of 15 fields matched the claim document. On the 6th, the insured number had two digits swapped: openIMIS found no member, the screen no longer matched the demonstration, and the program stopped before saving (no claim was written). The engine labels the 5 saved runs "Finished, not checked", because this program has no record check of its own yet.
- Same test copy, 12 Oct 2026: a test script typed one claim at a person's pace (about 39 words a minute, with 1 to 2.6 s pauses to read the document): 42.3 s from start to the Save click. A test script, not a person, and signed in with the blank form already open.
- OpenAdapt engine time per saved claim in the 12 Oct 2026 batch: 47.0 to 72.7 s, on a shared machine running openIMIS under emulation. OpenAdapt was not faster than the test script per claim.
- OpenAdapt on openIMIS 25.10 (open-source insurance system, not a US payer portal), July 2026, synthetic members: 3 of 3 active policies confirmed, 3 of 3 expired policies stopped, 0 active policies stopped by mistake. Checked by a read-only database query of the policy, product and dates.
- NAIC Insurance Data Security Model Law (#668): adopted in 28 jurisdictions as of 8 Oct 2026 (NAIC adoption map). NAIC Model Bulletin on the use of AI by insurers: adopted in 29 jurisdictions as of 8 Oct 2026.
- New York 23 NYCRR Part 500, section 500.11: third-party service provider security policy.
- OSFI Guideline B-10, third-party risk management, in force 1 May 2024; Guideline E-23, model risk management, in force 1 May 2027.
- US HHS guidance: third-party administrators of group health plans are generally business associates under HIPAA.
See it on your process
Bring one task your team types every day. We'll tell you whether it fits, and what a test on your own system would cover.