Today, I’m proud to announce Sardine AI Labs, an applied research group building fraud and financial crime AI models. As part of the launch, we’re funding 5 independent researchers with $375,000 in total funding. I’ve wanted to build this for a long time. Let me give you an early example of the work we’re doing: Typically, new card programs lack the data to detect fraud reliably, and they're often among the first targets for fraud rings. In testing on card issuers held out of training entirely, the very first foundation model from Sardine AI Labs improved fraud detection accuracy by 68% for a consumer card issuer and 41% for a business card issuer compared with standard machine learning approaches. Our consortium recently crossed 6.5B devices profiled and $1.8T in payments. This data, combined with our fantastic team of AI researchers and engineers, makes this lab and these early results possible. In addition, the Sardine AI Fellowship Program is an initiative I wanted to launch alongside Sardine AI Labs. When I was writing my PhD dissertation in the early 2000s, the preeminent research labs like Bell Labs, HP Labs, AT&T research did a lot to move research on various internet technologies forward. Over time, the concept of research labs withered away. This is our attempt to give back to the research community. We're funding up to 5 research fellowships, with $375,000 in total, for researchers who want to help push this work further. Key details: - Advanced-degree students, researchers, and faculty at US or Canadian universities. - Fellows work with the Sardine AI Labs team to build new foundation models, and are encouraged to publish in peer-reviewed venues. - Research areas include cross-institution generalization, multi-stream modeling, scaling laws, identity matching, real-time inference, cold-start learning, and applications beyond fraud. Here’s more about Sardine AI Labs and how to apply to the fellowship: sardine.ai/ai-labs I am absolutely thrilled to launch both of these initiatives. Thank you to the entire @sardine team who has brought this vision to life.

Behavior analysis is the next frontier of fraud prevention. As account takeovers, phishing, and other fraud vectors become more sophisticated, fraud prevention must move toward behavior analysis. The way you type is unique. It’s like how you talk. You have specific words you say more than others, a certain cadence you talk with, and maybe even an accent. On your phone (or laptop) it’s the same story — each person has a collection of traits that, together, make your style unique. That means if someone who isn’t you tries to log in, even with the correct info, that can be detected. @sardine has built a model that detects whether you’re the actual user or someone else just based on how you hold the phone. With that info, we can say whether it’s the actual user with 85% accuracy. That’s not definitive, but it’s a fantastic fraud signal. Especially combined with the thousands of other data points we have to detect fraudsters.

If a pickpocket steals your phone, they could steal all the money in your bank account. And in most cases, you won’t be refunded. Even if you have a PIN code and biometric authentication set up. How? A pickpocket does something called “shoulder surfing.” They stand behind you and watch you enter the PIN codes. Why does this work? 🤔 🐟🐟🐟 Most banking apps consider the PIN code a fallback (or even as more secure than biometric authentication). If biometric authentication fails, the user can use the PIN code to get full access to the account. Your computer may work like this, too. So the fraudster goes into your apps, fails the biometrics or just skips them, and enters your PIN code. Boom – they’re in. And once they’re in the account, they can do whatever they like. Including moving your life savings into another account they control. In many cases, you won’t be refunded. Why no refund? 🤔 🐟🐟🐟 To the bank, it looks like you authorized those transactions. Fraud investigators at the bank see legitimate-looking transactions securely approved. Fortunately, now there are some banks refunding users But there is more we can do… Every swipe, tap, and type you make is unique to you. If you suddenly started speaking with a French accent, people might notice. If we understand your unique behavior, we can notice those changes too. You have a specific way you type. A specific way you login. Even a specific way you hold your phone. At @sardine, we focus on the device and the behavior. So if any of those traits specific to you change, our fraud alerts start going off. That way, we’ll prevent more people from losing their life savings.

4/ For the next 24 hours, we’ve partnered with Ledger to waive transaction fees and rebates to celebrate the launch and let users experience the @sardine promise. Faster, safer payments for you and your users. Excited to partner with a company that shares so many of our values.

1/ Imagine a crypto on-ramp that’s the safest, fastest and has the highest conversion rates in the industry… It’s real, and it’s coming to @Ledger At Sardine, our crypto on-ramp is trusted by incredible companies and today we’re thrilled to add Ledger to that list of partners.