Latest / Fintech Conversations with Fexingo: Banking, Payments, and Financial Software Companies / How Fintechs Use Real-Time Identity Data for Instant Account Opening
Transcript
- Lucas: You know how opening a bank account or a trading account used to mean filling out a paper form, waiting days for verification, then mailing in a copy of your passport? That world is basically gone. The shift is being driven by real-time identity verification — pulling data from government databases, biometric facial matching, and document authentication in seconds. Luna: Right, and it's not just about convenience for the customer. For the fintech, it's about preventing fraud before it happens. If you can verify someone's identity instantly, you stop synthetic identity fraud at the door. Lucas: Exactly. And look, if you're finding these conversations useful for what you're building or running, the reason we can keep the show ad-free is listener support. If you wanted to help keep that going, you can buy me a coffee dot com slash fexingo. Luna: Yeah, every little bit helps us keep digging into these topics. Lucas: Alright, so back to identity verification — one of the most interesting recent developments is the use of liveness detection in biometrics to prevent deepfake attacks. The typical flow now: take a photo of your passport, then a selfie. The software checks the document's security features — watermarks, microprinting — and then compares your face to the photo. But the liveness check makes sure it's a real person, not a photo or a video. Luna: I've seen some apps ask you to blink or turn your head. That's part of liveness detection? Lucas: Yes, exactly. Passive liveness can analyze skin texture and subtle movements. Active liveness asks for an action. Both are evolving to catch ai generated deepfakes. A company like Revolut, which has millions of users across Europe, relies on these checks to onboard new customers in under two minutes. Luna: That requires a lot of data partnerships, right? With credit bureaus, government databases? Lucas: Absolutely. They integrate with services like Jumio, IDnow, or Veriff. These services check the document's authenticity by analyzing security features, then compare the selfie to the passport photo using AI. Some even read the NFC chip in newer passports for an extra layer of verification. Luna: And the cost? Is this only for big fintechs? Lucas: The pricing model has shifted to api based pay per check, so even startups can access enterprise-grade verification. Some providers charge as little as 50 cents per verification. That's a bargain compared to the cost of a fraud incident. Some fintechs report a 40 percent reduction in account opening fraud after implementing liveness detection. Luna: That's impressive. But there's also the question of bias. Some studies show facial recognition has higher error rates for people with darker skin tones. Lucas: It's a real concern. The industry is working on training data diversity, but it's not solved yet. Regulators are starting to look at this. The EU's AI Act includes provisions for high-risk AI systems, which includes biometric verification. So compliance is becoming more complex. Fintechs have to balance speed, security, and fairness. Luna: What about smaller fintechs that can't afford the compliance overhead? Are they locked out? Lucas: Not necessarily. Many verification providers handle compliance on their end, offering pre-built integrations that meet KYC and AML requirements. So a startup can plug in and be regulation-ready from day one. It's become a commodity. Luna: That makes sense. So what's the next frontier? Maybe decentralized identity where users hold their own verified credentials? Lucas: That's the holy grail — self-sovereign identity. Some fintechs are experimenting with blockchain-based identity, but it's still early. Standards like Verifiable Credentials are emerging, but adoption is slow. For now, the real-time, api driven model is winning. Luna: I can see behavioral biometrics becoming a layer on top — how you type, how you hold your phone. Continuous authentication. Lucas: Exactly. The initial one-time verification is just the start. The future is ongoing, passive authentication. We're already seeing it in some banking apps that monitor mouse movements or typing cadence. If the behavior changes, they lock the session. That's where it's heading. Luna: It's fascinating how far we've come from paper forms. And it's all happening in real time. Lucas: Right. And the key enabler is data availability — government APIs, network connectivity, cheap compute for AI. Those three things together made instant verification possible. It's a good reminder that fintech innovation is often about bundling existing tech in a smarter way. Luna: Well said. That is a wrap on real-time identity for opening accounts. Thanks for listening. Lucas: And if you want to keep the show ad-free, you know where to find us. Until next time.