You bring the book. We build the AI.
We're partnering with up to five MGAs to build the AI infrastructure and workflows your book runs on. We fund the build, you own the code, and you only pay when you use it.
Start applicationCohort 1 applications close Sep 8 — seats fill as we admit, so earlier is better.
Launchpad is an AI Insurance Accelerator from Dearborn Labs. We partner with a small cohort of the best MGAs, any line of business, to build the AI infrastructure and workflows they own. Partners pay nothing for the build; we earn only on measured outcomes like submissions triaged and claims closed. The inaugural cohort is limited to five MGAs, application-only.
A decade building and operating an AI-native carrier.
We founded and ran Clearcover from the data up. Operators and builders, forward deployed. Now we put that work behind a small group of MGAs and back their growth.
Published case studies: servicing, claims co-pilot, claims intake.
The upside is real. The build is the bottleneck.
You are busy running a great insurance company. You know AI can lift productivity and cut expense, but prioritizing, scoping, building, and implementing it is another full-time job. Let our forward-deployed engineers, who have built and run insurance for a decade, do that part for you.
Quote more of the funnel
Faster triage means more submissions priced.
Price sharper
Better risk selection protects the loss ratio your capacity partners watch.
Service leaner
Resolve the routine automatically. Keep the book you fought to win.
Close faster
Cut cycle time. Catch leakage earlier. Judgment stays with your people.
Built for the MGAs everyone wants to back.
The best MGAs are lean, fast, and deep in their niche. You're great at what you do, and you see opportunity for growth with technology, but rarely have the time or the capital to build AI on top of it. That is the gap we fund and build across.
- You run a sharp book of over $10M GWP, in any line of business.
- You can put real volume through the system, with an executive behind it.
- You are ready to get started.
- You do not need to know what to build yet. We can help with that.
Our team builds the AI infrastructure and workflows with our capital, and sticks around to make sure it succeeds.
Our forward-deployed engineers build the AI, embedded with your team. We earn only when you use it, on the measured outcomes we decide together. The risk sits with us.
We fund and build your AI infrastructure and workflows, and we carry the build risk. If you don't use it, you don't pay for it, and you own what we build.
We agree what success looks like, and how to measure it, together and up front. Then we earn on the work the system does: per submission triaged, per servicing event, per claim closed.
Not a capital project. We cover the build, so there's no cash payment before you start using it. You pay as the system does the work, and the cost scales with the work. It is an operating line, not a bet.
Start by answering a few questions. Define success together.
We agree the success metrics with you, in writing, before any build starts. Then we build, and we earn only on those results: per submission triaged, per servicing event, per claim closed.
Indicate interest
Briefly tell us about you and where you see AI helping.
We review and reach out
If it looks like a fit, we set up an interview and follow up with more questions.
Scope the work, define success
We scope the build together, and agree on what success looks like and how we measure it.
Build and implement
Our forward-deployed engineers build and implement it, embedded with your team to the extent you want.
We stick around
We stay after go-live to make sure it works and keeps working for you.
We build it. You own it.
The AI infrastructure and workflows we build are yours to keep. No license, no subscription. We fund the build and earn only on results you can measure. Same side of the table, same definition of success.
An accelerator that backs you. We fund and build your AI infrastructure and workflows. You own the result.
Our team. Our track record. Behind your book.
Start applicationQuestions, answered
What does it actually cost?+−
There are no upfront costs. Prior to any build, we'll define success and the metrics that measure it. We'll then tie outcome-based pricing to those metrics. Pricing may differ depending on the workflow, but terms will always be agreed upon upfront, measurable, and based on output. No surprises, ever. We anticipate a range of pricing models, such as price-per-event, increase in premium, or a capture of efficiencies made.
What do you need from us?+−
Real opportunity to grow together through custom AI technology. We expect there to be an executive sponsor, access to the systems the workflows touch, and real volume to run through.
Does it work with our systems?+−
Yes. We build into your existing stack: policy admin, carrier, and TPA systems. No rip and replace.
How long does it take?+−
We find that usually procurement is the longest step in the process. We can get started right away with the right partners, and build time can vary depending on the scope of work, starting around a few weeks.
Will this replace our team?+−
No. We automate the repetitive work so your team spends time where judgment matters. At the carrier we built, 92%+ of adjusters chose the tools because they made the job better. We can work alongside your tech and IT teams, or as a standalone group.
Are we locked in?+−
No. You only use what works. You only pay for what you use. No upfront cost to claw back, no obligation to keep a workflow that stops earning its place. Full cohort terms shared during selection.
Is this a capital project or an operating cost?+−
Operating. We cover the build. Your only cost is variable, tied to the work the system does.
Why is acceptance limited?+−
We do the building, and doing it well means working closely with each partner. A small cohort means every MGA we take on sees real results.
When do applications close?+−
Cohort 1 applications close Sep 8, but we admit on a rolling basis and seats fill as we go. Up to 5 seats, application-only.
You bring the book. We build the AI.
Cohort 1 is up to 5 seats. Applications close Sep 8 — and seats fill as we admit, so earlier is better.
Start applicationYou owe nothing until it works. Confidential.