Conor Burke spent much of his career in the back office of a major bank in Ireland. His team was tasked with digitizing the onboarding process, especially the manual, document-heavy review workflows that cost the bank millions of dollars every year, and to avoid solving fraud. He said the biggest challenge was figuring out how to remove the human element without compromising risk and fraud controls.
Inspired by this, Burke and his twin brother, Ronan, launched Burke Register, an AI-powered document fraud detection service. Built for fraud, risk, and operations teams in the fintech and finance industries, Inscribe uses AI trained on hundreds of millions of data points to return results, says Ronan.
“Tedious document checks create friction in the account opening and underwriting processes, but automation alone is not the answer,” Ronan told australiabusinessblog.com in an email interview. “We believe that automation without fraud detection is reckless. That’s why Inscribe is the complete package that helps companies detect fraud, automate processes and understand creditworthiness so they can approve more customers faster.”
Inscribe parses, classifies and matches the data of financial onboarding documents, highlighting any discrepancies between the provided documents and the recovered documents using the AI-powered fraud detection. Document details, including names, addresses and bank statement transactions, are automatically digitized to generate individual customer risk profiles with snapshots of bank statements and transactions.
Last September, Inscribe rolled out a credit analysis and bank statement automation component that provides most of the data needed to make credit decisions, including bank statement cash flow details, transaction parsing, and paycheck parsing. Ronan claims that Inscribe can extract and return important details including names, addresses, dates, transactions, and salaries in seconds.
In the features it offers, Inscribe is comparable to many of the other anti-fraud tools out there, such as Resistant AI (which grossed $16.6 million in October 2021) and Smile Identity (which grossed $7 million in July of that same year) . However, Ronan claims that what sets it apart is its AI-first approach, which relies on original data collected through previous partnerships with clients.
“We had seen fraud detection and document automation companies in our space trying to build a perfect solution right away without talking to customers, but they had since shut down. They were unable to overcome the cold start problem; they weren’t able to build a product from the ground up because they didn’t have access to the data their customers were using,” said Ronan. “This comes back to the first rule of machine learning: start with data, not machine learning. If you don’t have a good dataset, you’re wasting your time. You end up choosing the wrong model or training a model on data that doesn’t perform as you expect.”
AI is far from perfect – history has shown that much is true. For example, during the pandemic, there were fraud detection systems that signal abnormal behavior confused by new shopping and spending habits. Elsewhere, automated algorithms designed to detect social fraud have proven error-prone and designed in a way that essentially punishes the poor for being poor.
But beyond the veracity of Ronan’s claims, there’s clearly something about Inscribe’s platform that draws in high-profile customers. TripActions, Ramp, Bluevine, and Shift are among the startup’s clients.
Investors, in turn, have won. This week, Inscribe closed a $25 million Series B funding round led by Threshold Ventures with participation from Crosslink Capital, Foundry, Uncork Capital, Box co-founder Dillon Smith, and Intercom co-founder Des Traynor. The infusion brings the total amount the startup has raised to date to $38 million, including a $10.5 million Series A round that closed in April 2021.
Perhaps it’s the relative ease with which Inscribe’s solution can be implemented. As Ronan rightly points out, Inscribe solves the problem of building an in-house fraud detection solution or hiring a large data science team.
“AI and machine learning models take advantage of as much data as possible, but each individual company is limited to just their own data set. So a homegrown solution just can’t be as effective as one that draws from countless data sources,” said Ronan. “That’s why companies are instead partnering with document fraud detection solutions—criminals commit fraud in a variety of ways, and those solutions pull data from across their customer base to more quickly identify coordinated attacks and emerging trends.”
Instilling fear probably helps too. A recent one survey suggests that the average US fintech loses $51 million to fraud every year, a statistic Ronan quoted me during our interview.
“An increasingly digital, geographically dispersed and fast-paced world makes it harder than ever to know who you’re doing business with, leaving businesses unsure about which potential customers are trustworthy,” said Ronan. “Fintechs have been able to build an online world, but traditional financial institutions face the challenge of moving away from legacy systems and embracing true digital transformation. And they need to do it all while reducing fraud and friction to have competitive customer experiences.”
When asked about expansion plans, Ronan says Inscribe will likely double the size of its 50-person workforce in the next 12 to 18 months.