Inspired By Own Experiences, Israeli Startup Tackles Fraud Using Video ID

An Israeli security startup inspired by its co-founder’s real-life experiences of fraud, wants to bring identity verification to susceptible industries like online dating.

I Am Real uses big data processing algorithms to ensure identity confirmation, combining “analysis of a person’s social activities and real-time image processing”, to determine if they are who they say they are.

Founded in 2013 by Nadav Benedek and Shmulik Primes, I Am Real has two main features: its Social Network Validation and Video Face Validation.

The former uses Facebook Connect, pulling a wealth of data from the social network, analysing it, and sending the results back to the user.

Speaking about this social validation, co-founder Nadav Benedek told GDI: “Over the years, we’ve seen and analysed many fake profiles – profiles that present facts that are different from reality. Some profiles were created in order to con other people, some for marketing purposes, some for criminal activity, and some as a means of sabotaging a platform.

“We discovered that these malicious profiles have different patterns, or sometimes lack patterns that genuine profiles have. By learning those patterns, we can assess the probability that a given profile is fake or not – and our live tests show that it works.”

The Israeli startup’s other security system is its patent-pending video verification feature.

This requires users to record a short video of themselves saying a randomly generated code.

Using computerised technology, I Am Real then uses this video to match it with images from their social network profiles, or their passport/driving license, if the service is being used for something like online lending, and requires such documentation.

Benedek said once I Am Real has the video, “our algorithms can not only validate their identity, but can also tell if they are using a picture that is true to what they look like today. This is especially relevant for dating services, where some people might use old or touched-up photos.”

The idea behind the security company came after Benedek fell victim to an online fraud himself, when looking online for cheaper alternatives when importing goods from Brazil.

He said: “I went on Alibaba, and found another supplier, with better prices. I contacted this supplier, and started to negotiate with him. We spoke on the phone (they had perfect English), and finalised a shipment. They even sent me photos of their hangar.

“Then, after we agreed on all the details, we signed a contract, and I made the payment. Right after the payment went through – they disappeared, as if they fell off the face of the earth. Apparently they sent me fake photos, and the phone number was of a temporary sim card.

“If I (or Alibaba) had a service like I Am Real back then – it would have probably never happened.”

And the company is looking to integrate its service with online dating clients, to help prove user identities, and cut out the fraud that takes place on some sites.

In terms of positives for potential online dating partners, co-founder Benedek said dating brands can expect:

• An increase in engagement between users, because they trust each other more.
• An increase in the site and brand’s credibility.
• Less fear of online scams, identity theft, paedophilia and other criminal behvaior. Reduced fear of Stranger Danger.
• A decrease in the legal risk the platform faces.
• A decreased risk of negative PR for the platform.

In terms of implementation, Benedek said the process was simply “a matter of adding a few lines of code” to a client’s platform.

In addition to the verification service, I Am Real boasts that it also generates a “Social Score” for users, which could inform companies whether an individual is likely to become a paying customer.

Benedek explained to us how the feature works: “We give a feedback API for the platform, in which the platform can report back to us if the customer has converted to a paying customer or not.

“After the first 1,000 users have registered to the system, we learn from each one of them if they became paying customers or not.

“Then, we automatically create a machine learning prediction model that can predict, especially for that specific platform, based on the massive social data the user has, if they will convert or not.

“Then, when the 1,001st user arrives, immediately after he’s done the Facebook connectivity during sign up, we can predict if he’s likely to convert or not, exactly like Waze can predict how long it will take you to drive from point A to B, based on past experience of other drivers.”

The platform is available in four package deals, ranging from “free” to “gold” standard.

When using the “free” package, the first 100 users come free of charge, and prices rise according to the number of services required by each customer.

For more information about integrating with I Am Real’s API, check out the API docs on the company’s site.