# Inscribe AI

> Inscribe AI sells document-fraud detection to banks, credit unions, fintechs, and lenders. Its software inspects uploaded bank statements, pay stubs, tax forms, IDs, and business records with a layered mix of forensic, network, semantic, perceptual, and public-data checks, then gives risk teams an explainable assessment. Founded in 2017 after its creators saw how much supposedly digital finance still depended on manual review, Inscribe has raised about $38 million, survived a sharp 2024 reset, and repositioned its product around AI agents that investigate whole applications rather than merely extracting fields from individual files.

- **Founded:** 2017
- **Headquarters:** San Francisco, United States
- **Founders:** Ronan Burke (Co-founder and CEO), Conor Burke (Co-founder and CTO), Oisín Moran (Founding team / co-founder reported in 2021 coverage; current role not public), James Eggers (Founding team / co-founder reported in 2021 coverage; current public title not confirmed)
- **Team size:** Approximately 35-44 people; Y Combinator listed 35 and supplied company data listed 44 in 2026, while LinkedIn's broad band is 51-200.
- **Products:** Document Fraud Detection, Credit Analysis and Bank Statement Automation, AI Risk Agents, Applicant-Level Fraud View, Inscribe Assistant
- **Notable:** Accepted into Y Combinator's Summer 2018 batch., Ronan and Conor Burke were named to Forbes 30 Under 30 Europe in 2020., Reported detecting more than $40 million in fraud per month for customers by April 2021.

## Products & services

- **Document Fraud Detection** — Analyzes bank statements, pay stubs, tax forms, IDs, invoices, business filings, and other submitted documents for manipulation, fabrication, reuse, and AI generation using layered forensic, network, semantic, perceptual, and metadata signals.
- **Credit Analysis and Bank Statement Automation** — Extracts transactions and income details, categorizes cash flow, and prepares credit insights for underwriting workflows.
- **AI Risk Agents** — Pre-trained agents for onboarding and underwriting that combine proprietary detectors, third-party data, code execution, objectives, reasoning, and feedback loops to complete multi-step reviews.
- **Applicant-Level Fraud View** — Connects evidence across the complete application rather than reviewing each uploaded document as an isolated object.
- **Inscribe Assistant** — A conversational investigation layer inside the web application that explains signals, compares documents, examines transactions, and answers questions grounded in a customer case.
- **Custom Transaction Insights and Configurable Decision Engine** — Lets institutions define, backtest, and deploy their own transaction checks, tune detector severity, manage blocklists, and align decisions with internal risk policy.
- **REST API and Web Application** — Customers can upload and review documents through Inscribe's web interface or integrate customer folders, document uploads, results, and webhooks into existing workflows by API.

## Achievements

- Accepted into Y Combinator's Summer 2018 batch.
- Ronan and Conor Burke were named to Forbes 30 Under 30 Europe in 2020.
- Reported detecting more than $40 million in fraud per month for customers by April 2021.
- Raised a $25 million Series B in January 2023, bringing reported total funding to about $38 million.
- Plaid reported document review in about 30 seconds, down from one to two days, with half of reviews automated.
- Logix Federal Credit Union reported more than $3 million in potential fraud losses prevented within eight months.
- BHG Financial reported reviews more than 90% faster and millions in potential fraud losses prevented.
- BCU has publicly attributed multi-million-dollar fraud-ring prevention to Inscribe detections.
- Maintains SOC 2 Type II and ISO 27001 certifications.

## Latest updates

- **2024-01** — Inscribe reduced headcount by just under 40%, mostly in go-to-market and operations, after missing revenue goals amid higher rates and a weaker fintech market.
- **2024-09** — Launched AI Risk Agents for multi-step onboarding and underwriting workflows.
- **2025** — Expanded from document-level signals toward an applicant-level view that correlates evidence across files and cases.
- **2026-06** — Released the Inscribe Assistant, Custom Transaction Insights, a Configurable Decision Engine, a Taktile integration, and dark mode.
- **2026-07** — AWS published the architecture behind Inscribe's Bedrock-based multi-model system, reporting under-90-second investigations and a 20x speed improvement over manual review.

## Links

- Website: https://www.inscribe.ai/
- LinkedIn: https://www.linkedin.com/company/inscribeai/
- Twitter/X: https://twitter.com/inscribeai
- YouTube: https://www.youtube.com/@inscribeai

---

Profile page: https://yespress.io/inscribe-ai
Published by YesPress — https://yespress.io
Last updated: 2026-08-20