# Mighty - Check Customer Files Before People or AI Act

> Mighty checks customer-submitted documents, images, OCR output, and extracted
> text before reviewers, AI tools, OCR, or automation rely on them.

Mighty is designed for production workflows that process outside-submitted
material. The first wedge is P&C insurance, but the control point applies more
broadly anywhere customer or vendor files can steer a decision.

## Quick answers

- One endpoint: `POST /v1/scan`
- Inputs: text, images, PDFs, documents, OCR output, and extracted text
- Output: allow, review, or block with evidence and audit-friendly identifiers
- Deployment: hosted API, hashed-input mode for sensitive workflows, and
  private-cloud / BYOC / on-prem discussions for stricter environments

## What Mighty does

Mighty is used before downstream systems act on submitted material. It can scan
text, images, PDFs, documents, OCR output, and extracted text through
`POST /v1/scan` and return risk signals, extracted text when available, and an
allow/review/block workflow action.

## Positioning

Mighty is not an IDP replacement and does not claim to validate every fraud
category. The tighter claim is simpler: check risky customer files before
people or AI act on them.

The first wedge is P&C input trust before workflow action:

- Claims intake before adjuster review
- File screening before OCR and document extraction
- Repair estimate, damage photo, and vendor invoice review
- Extracted text inspection before AI or automated tools use it

## Sourced Market Numbers

- NAIC cites the Coalition Against Insurance Fraud estimate that fraud costs
  businesses and consumers $308.6 billion per year across insurance lines.
- NAIC cites a $45 billion property and casualty fraud estimate.
- NAIC cites the FBI estimate that insurance fraud costs the average family
  $400 to $700 per year in premiums.

Source: https://content.naic.org/insurance-topics/insurance-fraud

## Public pages

- [Home](https://trymighty.ai): Product overview and positioning
- [Docs](https://trymighty.ai/docs): Quickstart for `POST /v1/scan`, request
  fields, and response fields
- [Product](https://trymighty.ai/product): Product overview and workflow fit
- [Use Cases](https://trymighty.ai/use-cases): Claim packets, repair estimates,
  damage photos, invoices, OCR output, and adjacent workflows
- [Pricing](https://trymighty.ai/pricing): Public SCU pricing and deployment
  options
- [Why Mighty](https://trymighty.ai/why-mighty): Why input trust matters before
  workflow action
- [Security](https://trymighty.ai/security): Data handling and security
  practices

## Optional

- [Privacy](https://trymighty.ai/privacy): Hosted service privacy policy
- [Terms](https://trymighty.ai/terms): Hosted API and dashboard terms

## Contact

- Website: https://trymighty.ai
- Email: hi@trymighty.ai