The European unicorn using machine learning to make the internet a safer place.

Michael Butler, ML Community Lead
Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.

The Story of Veriff

Not too long ago, a teenager in Estonia was doing all he could to help out on his family’s farm. One day, he needed to order some supplies for the farm, but was rejected from setting up a payments account because he wasn’t yet 18 years old. Without thinking twice, he tampered with the photo of his government issued ID card and changed the year of his birth from 1994 to 1984.

Just like that, his payments account was activated. But the thought of how easy it was to trick the system was unsettling for Kaarel Kotkas. In 2015, Kaarel founded Veriff to bring the slow and outdated identity verification process into the 21st century.

The Veriff team quickly realized that traditional identity verification methods, such as asking for a passport or driver's license, were not enough to keep up with the growing number of digital transactions. They knew they needed to build a more advanced solution employing machine learning models built by data scientists and machine learning engineers that could verify identities in real-time, all while providing a seamless user experience.

"We want to solve the crisis of identity online. You shouldn’t be able to pretend to be someone else and create chaos. If we solve that, it would be huge."

- Kaarel Kotkas, CEO

Using Machine Learning to Disrupt Identity Verification

Think about how much of your own life is online. On any given day you might use an online service to open a bank account, sign up for a new stock trading app, or even register for a new dating service. For most companies, it is more important than ever to be able to verify the identity of people online. As a consumer, it’s something you should want these companies to care about as well! Fraudsters are constantly finding new ways to exploit vulnerabilities in the system, and both businesses and consumers need to be able to protect themselves from these threats.

That's where Veriff comes in: An AI-powered identity verification company that helps businesses verify the identities of their customers in a quick, accurate and secure way. Veriff's services are used by a variety of businesses, including banks, financial institutions, gaming companies, and e-commerce retailers.

  • Veriff's technology is based on a combination of artificial intelligence, machine learning and biometrics. This allows Veriff to verify a user's identity in a matter of minutes, all while still maintaining a high level of accuracy.
  • The technology doesn’t just scan a document. It requires an end user to record a video of themself holding their ID so that Veriff’s proprietary ML models can scan thousands of data points.
  • In addition to its accuracy, the technology is also incredibly scalable. It can verify the identities of users from all over the world, in a variety of languages. This makes Veriff an incredibly valuable tool for any business that wants to expand their reach into new markets.
Veriff's identity verification flow

In order to build all of this, Veriff needed to hire a world class team of data scientists and machine learning engineers to work together to build, train, and deploy machine learning models into their product. They already have over 15 members on the team.

Veriff’s Competitive Advantage

When reimagining identity verification, they took a new and unique approach:

  • Users have to show both the ID and their face at the same time.
  • Veriff focuses on a video first approach.
  • With video, they can analyze on a frame by frame basis. That means their machine learning models can then analyze up to 2,000 data points for each verification.
  • Veriff can be 30x more accurate and 6x faster than their nearest competitors.

How Veriff is Truly Changing the World

But Veriff is not just a tool for businesses. It truly is a tool for changing the world. By making it easier for businesses to verify the identities of their customers, Veriff is helping to reduce fraud and make the internet a safer place. Here’s a few real world examples:

  • Protecting children from online predators: Veriff's technology can be used to verify the age of users on social media platforms and gaming sites. This helps to prevent children from being exposed to harmful content and from being targeted by predators.
  • Helping to prevent financial fraud: Veriff's technology can be used to verify the identities of users who are opening bank accounts or making online payments. This helps to prevent criminals from using stolen identities to commit fraud.
Veriff compares live video and IDs to verify identities

How can you use Veriff for your business?

There are number of different use cases for Veriff, depending on the type of business you’re operating. Here are a few ideas:

  • Verify the identities of users on your website or app. This will help you to protect your users from fraud and to comply with KYC regulations.
  • Verify the identities of employees or contractors. This will help you to ensure that you are hiring the right people and that you are complying with employment law.
  • Verify the identities of customers who are applying for loans or other financial products. This will help you to reduce your risk of fraud and to make sure that your customers are who they say they are.

With backing from investors such as Accel, IVP, Tiger, Taavet Hinrikus, Ashton Kutcher, and Paul Buchheit, and having received unicorn status, they’re well positioned to achieve their mission of increasing transparency online and making the internet a safer place.

A note from the Modelbit team

Hi there! We love to learn and write about companies building incredibly cool products with machine learning. We aren’t paid or asked to write these articles. We just think it’s important to highlight great work in the ML space.

Want to suggest a correction or a company using ML that we should highlight? Get in touch at!

Deploy Custom ML Models to Production with Modelbit

Join other world class machine learning teams deploying customized machine learning models to REST Endpoints.
Get Started for Free