Automate PII removal from data

Automate PII removal from data

Identify and remove PII from your data with Tray.io and AWS Comprehend

We explore a crucial topic for businesses: identifying and removing Personal Identifiable Information (PII) from data. This topic and approach is especially useful today when working with Generative AI API: ChatGPT and we might not want to send to OpenAI our sensitive data.

Learn about the reasons businesses need to tackle this issue, including compliance with regulations like GDPR and ensuring data security. Discover a powerful solution leveraging our partner Tray.io, an iPaaS and automation platform, and Amazon AWS Comprehend service API.

PII removal with Tray.io and AWS Comprehend API by Naonis

Misplaced sensitive data?

  • Have you ever seen data stored in the wrong place within your Customer Relationship Management software (CRM), or any other internal system?
  • Has someone within your business seen unauthorised sensitive information due to human error?
  • Does your business fully trust third party APIs with your customer’s PII data? Some recent examples include the ubiquitous Generative AI API: ChatGPT where PII might be leveraged to train further the AI models

Tray.io + Amazon Comprehend API = removed PII!

Now you have a solution: an automated and elegant mechanism to detect and redact PII data. An automated second set of eyes.

We demonstrate a step-by-step process, guiding you through identifying, removing, and safely storing PII data. This essential solution can help you with compliance, data security, and trust.

Watch the video for further details!

Key points covered

You can also access the solution for free as a Tray.io snippet right here!, and if you are not yet a Tray.io customer and are interested in becoming one contact us!

Do you want to discuss your Tray.io needs?

Getting help is super simple: you can reach out for an initial conversation about how automation and integration can be beneficial to increase efficiency within your business.

Frequently Asked Questions

Why is it essential to remove Personal Identifiable Information (PII) from business data?

Removing PII is crucial for compliance with regulations like GDPR, ensuring data security, and avoiding unauthorised access or exposure of sensitive information.

A tangible example covered in the video to help limit the potential exposure of sensitive information is to redact personally identifiable information before calling an API.

In our case we referred to OpenAI’s ChatGPT generative AI API, showcasing the practical importance of PII removal in real-world common scenarios.

How can businesses identify and remove PII from their data?

The video demonstrates a solution using Tray.io, an iPaaS and automation platform, along with AWS Comprehend service API.

This combination provides an automated mechanism to detect, redact, and then safely store PII data.

Why choose AWS Comprehend for PII removal?

AWS Comprehend is chosen for its readiness, availability with AWS accounts, the fact that most SaaS software available today is hosted already in AWS (and therefore the data contained within those softwares), and finally the assurance that Comprehend does not use input/output text for its AI model’s training.

How is the PII removal workflow structured in Tray.io?

The workflow involves generating a demo data set, passing it to AWS Comprehend for PII identification, removing identified data using JavaScript code, and potentially restoring the data if needed.

The video provides a detailed overview of the workflow components.

What are the benefits of using Tray.io and AWS Comprehend for PII removal?

The combination of Tray.io and AWS Comprehend provides businesses with an automated and elegant mechanism to detect, redact, and securely store PII data.

This solution helps ensure compliance, enhance data security, and build trust.

We also collaborated with Tray.io to release the snippet used for this demo for public consumption, so that you can use it too!

All trademarks mentioned on this page are the property of their respective owners. The mention of any company, product or service does not imply their endorsement.