**Resistant AI** provides in-line prevention and off-line detection services that protect AI algorithms from manipulation and attacks.

The solution we provide for RPA and document processing workflows protects the automated systems from manipulations by submission of forged, modified or otherwise manipulated documents.

efc5a8a1-d168-47bf-b6fc-af311a3dc205.jpg

Head of Solutions

Developer

**What documents are typically processed?**

Typical documents include invoices, payroll slips, bank statements, KYC documents, etc. 

  

**What are the main capabilities of this component?**

Resistant Documents detects forgeries and modifications related to identity changes, fraudulent modification of account numbers and transaction details.

  

**What do I need before I start using it?**

In order to use this activity you have to pass both your _API key_ and a _path to a document_ as input arguments. The activity then returns a _response type_ and a _response message_.

  

**What responses can I expect?**

There are five response types (_TRUSTED, NORMAL, WARNING, HIGH RISK_ and _EXCEPTION_) reflecting a confidence level, except the last one.

  

**What does the response message consist of?**

The response message consists of a JSON string, so it can be deserialized afterward.

Resistant Documents protects automated processes that rely on unauthenticated documents received from third parties in PDF, JPEG and other image formats.