> For the complete documentation index, see [llms.txt](https://propscreen.gitbook.io/propscreen/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://propscreen.gitbook.io/propscreen/introduction/solution-overview/how-propscreen-addresses-the-problem.md).

# How Propscreen Addresses the Problem

PropScreen is a dynamic tool specifically designed to safeguard organizational sensitive information generated by LLM outputs. Here’s how it works:

1. Multi-Layered Screening: It employs a three-tiered approach to identify and block sensitive information. Responses are checked against general personal information patterns, a customizable list of trigger words, and a database of sensitive data hashes.
2. Efficient and Cost-Effective: Unlike solutions that require costly and time-consuming model retraining, PropScreen uses a sensitive information context scanner and a keyword database to minimize computational load and efficiently protect data.
3. Enhanced AI Governance: By adding an extra layer of security and control, PropScreen strengthens existing AI governance programs and helps maintain a competitive advantage by safeguarding proprietary information and trade secrets.

In the near future, PropScreen will support Role-Based Access Control. PropScreen will assess the user’s role to ensure that data sharing is appropriate and limited to those with a need-to-know.