Earlier this month, members of our Lab team participated in the DisruptAI Hackathon, hosted by Havas in Toronto. The challenge of the Hackathon was about using AI and Data Science to disrupt an industry.
Before brainstorming, the team thought about what they wanted out of the Hackathon. First was hands-on experience with Microsoft’s ML technologies and an understanding of what it takes to make those technologies work. Second, the team wanted to stay within the domain of Interac operations, meaning e-Transfer, Digital ID, Flash, etc. And last, they wanted a solution that was a natural extension of problem statements that the Lab team had been exposed to over the last year, so the prototype developed would be tangible to a real problem.
While coming up with ideas the team worked to come up with a solution that could leverage available ML technology to provide some verification features of a Digital ID solution. This integration could be used to help reduce fraud and increase limits on digital transactions.
With this framework in mind, the team landed on creating a build that uses biometrics: “Biometrics are a fairly developed way to provide this,” says Vlado Jokic, Technical Lab Manager. “So we decided to experiment with Microsoft’s face and voice recognition features, with fairly positive results.”
In the end the team built a solution tentatively named p.AI.d: Paid with Artificial Intelligence Imbedded. Canadians are familiar with sending Interac e-transfers, but usually to people they know and trust. Non-transferable personal identification metrics can be deployed with AI to send larger e-Transfer amounts while making these exchanges more secure. Geo location services, repeat pattern machine-learning (ML) and data can also be analyzed to reduce fraud.
Now that the Hackathon event is over, the team is working on integrating more Azure features into an sample bank (called an ‘FI emulator’) that the Lab has also worked on. This integration allows the Lab to demonstrate digital transactions from end to end, using the emulator to show the experience for people on the end of each transaction. They’ll also be working to clean up the code that was written in the 24-hour Hackathon, and demoing at Interac HQ.
With updated iterations of p.AI.d, the team at the Lab have completed a demo at Interac headquarters to show how partners can use the solution and its potential for optimizing digital payments.