In partnership with

Hey, fintech fam πŸ’œ

We are officially 8 weeks away from FTW: San Francisco. And if I could wish for one thing, it would be that you join me.

At our last summit, an Academy of Fintech member closed her first deal.

Another attendee got a job.

Another met her first-ever investor in the room.

That’s why we build FTW.

I know the AI conversation can feel overplayed right now. But I still believe AI has the power to fundamentally improve financial servicesβ€”if we focus less on the hype and more on what actually works.

So in San Francisco, we’re bringing together the executives and builders actually putting AI to work at companies like Chime, Grasshopper, Candidly, Tabapay, and more to unpack what’s working, what isn’t, and the tactical moves leaders can take back to their businesses.

Because AI becomes meaningful when it improves someone’s financial life.

Early Bird tickets are only available through August 14. If you’ve been thinking about joining us, now’s the time.

Let’s get into it. ✨

#SPONSORED

4x your communication output. Same quality. No burnout.

The bottleneck isn't what you want to say β€” it's how long it takes to type it. Wispr Flow removes the bottleneck.

Speak naturally and get polished, send-ready text for executive summaries, client updates, board recaps, investor notes, or just the 30 Slack messages you're behind on. Flow strips filler, formats numbers and lists, and preserves your tone.

Used by teams at OpenAI, Vercel, and Clay. 89% of messages sent with zero edits. Works in every app on Mac, Windows, and iPhone.

#TRENDING

Every Thursday, I break down the fintech stories that matter most β€” grounded in my reporting, interviews with industry leaders, and what I’m seeing unfold across the industry.

#1 Capability Isn’t the Problem: What the Plaid-Sierra Deal Tells Us About Agentic AI in Finance

Source: Plaid

Plaid announced this week that customers can now securely connect their bank accounts with Sierra's AI agents directly inside a conversation, using Plaid Link.

The idea is that an agent doesn't just talk to a customer.

It can request permission to access their bank account mid-conversation and use that information to actually finish a task, like refinancing a loan or resolving a fraud claim.

That’s agentic AI performing at the highest level in fintech. Chatbots are cool, but an AI agent who can take action on your behalf is even cooler.

What This Looks Like in Practice

Take a loan application.

If a borrower falls just short of qualifying based on their credit score, the process typically stalls there; the file sits, the applicant waits, and the conversation with whatever agent or advisor is helping them effectively ends.

That's the exact scenario Plaid and Sierra used to explain why this matters in practice, and it's a good one, because it's the moment every fintech operator I interview says agentic AI is supposed to fix and hasn't yet.

In this version, the agent doesn't stop when the credit score becomes a hurdle.

It requests permission to look at the applicant's cash flow instead, and uses income and spending patterns to keep the application moving toward an approval.

Source: Pexels

Capability is the Easy Part

Every conversation comes back to this: agents are only as useful as the data they can reach, and the data that would make them useful is the data people are most afraid to hand over.

You’ve heard me talk about why trust matters so many times this year.

That’s because in fintech, you almost can’t go forward with deploying AI in any part of your business without the trust layer being there first.

Will Robinson, Plaid's CTO, and Clay Bavor, Sierra's co-founder, addressed that directly in announcing the deal.

β€œFinancial conversations require a higher bar than most AI interactions," they said in a statement.

If you take away anything from today’s newsletter and any newsletter I write about trust, let it be that. Financial conversations require a higher bar than most AI interactions.

Robinson and Bavor also wrote that the integration is built around "security, consumer control, and human oversight," with every interaction starting with explicit consumer permission through Plaid.

Why It Matters

We already know AI agents can reason well enough to handle a multi-step workflow. That’s not the bottleneck.

The bottleneck is getting people to connect their bank accounts in the first place.

Plaid and Sierra are effectively building the consent infrastructure that has to exist before any of the more ambitious agentic use cases, like autonomous underwriting or continuous fraud monitoring, can move past pilot programs.

Fintech has spent a decade earning the right to hold people's financial data, and that trust, not the underlying model, is the actual advantage the industry brings to the agentic AI conversation.

That’s the argument I'm building FTW: San Francisco around this fall. Early bird tickets are on sale now until August 15th!

#2 Visa Just Bought the Other Half of Trust

Visa announced this week that it's acquiring BioCatch, the behavioral fraud detection company, for $2.4 billion in cash.

BioCatch works by analyzing thousands of signals, such as keystrokes, touch gestures, and device handling, to determine in real time whether the person on the other end of a digital banking session is who they say they are.

It already protects 760 million users across more than 350 banks globally, and Visa is buying it specifically because account takeovers and scams now cost the global economy over $1 trillion a year, a number that's climbing as AI makes those attacks easier to run at scale.

β€œReal-time insights into customer intent continue to grow increasingly essential for institutions to establish trust within digital banking sessions,” said Gadi Mazor, the CEO of BioCatch.

β€œFor more than a decade, we've demonstrated behavior’s unique ability to distinguish the criminal from the legitimate.

Visa already has a proof point for what that looks like at scale.

Its Decision Manager platform screened 3.2 billion transactions in 2023 and prevented an estimated $33 billion in fraud losses, with 98.7% of those transactions resolved automatically by AI, drawing on a network that processes over 269 billion transactions a year.

Here’s the truth: the fintech industry can build all the consent infrastructure it wants, but permission only means something if the person granting it can trust the system on the other side to actually protect them once they've said yes.

Why It Matters

Agentic AI in financial services only scales if fraud prevention scales with it.

Every new connection point between a customer and an agent is also a new attack surface, and fraudsters are already using generative AI to build synthetic identities and deepfakes convincing enough to pass as a real customer.

It's the exact conversation Frances Zelazny is anchoring at FTW: San Francisco's Security Summit on October 1: "innovation without security doesn't scale, and in the agentic AI era, the stakes are higher than ever."

#3 FIS CEO Stephanie Ferris Says Finance’s Real Moat Isn’t the AI Model

I sat down with FIS CEO Stephanie Ferris at the FIS Emerald Summit in Orlando for Humans of Fintech.

During our conversation, she pushed back on the idea that the AI model itself is where the competitive advantage lives.

Security, compliance, and auditability are the actual barriers to entry in finance, not whichever model a bank happens to plug in, argues Ferris.

FIS's strategic partnership with Anthropic, which was announced at the summit, is built around that split.

Ferris also talked about what she's calling "orchestrated intelligence," where AI moves from answering individual questions to running full agentic workflows.

The first one FIS has deployed is in financial crime detection, automating the data collection fraud teams used to do by hand, cutting that manual workload by up to 90% and reducing processing time by roughly half.

A second workflow, through a new partnership with InvestCloud, automates note-taking and meeting prep for wealth advisors so they can spend more time actually advising.

FIS is positioning this as something smaller community banks can access too, not just the largest institutions, through its Flex platform.

Why It Matters

Ferris closed our conversation with the idea that trust is the essential infrastructure of the AI economy, and that compliance and security aren't friction points anymore; they're the foundation for growth.

That's the same infrastructure question running through this week's Plaid-Sierra and Visa-BioCatch news: the model matters, but it needs the systems built to make it safe to use.

FTW: SF IS ONLY 8 WEEKS AWAY

Drew and I look like we’re about to drop the coolest rap album of all time πŸ˜†

  • Brex is investing in agentic AI

  • Chime is investing in agentic AI

  • Stripe is investing in agentic AI

  • Plaid is investing in agentic AI

The list goes on. Don’t miss meeting with 500+ other leaders in fintech and learn how your fintech should use agentic AI.

Early bird tickets end August 14th. Get yours today! 🐦

FINTUNES

LET’S CONNECT

πŸ“° Share this newsletter with a friend and start growing your network.

πŸ”— Connect with me on LinkedIn for daily insights on female leadership.

🀝 Grow your business through content & community by partnering with me.

πŸ“£ Promote yourself to 50,000 subscribers by sponsoring this newsletter.

🎀 Host an epic event by booking me as a speaker, moderator, or emcee.

πŸ“š Increase your expertise by ordering your copy of my book, Fintech Feminists: Increasing Inclusion, Redefining Innovation, and Changing the Future for Women Around the World.

That wraps up today’s editionβ€”thanks for reading! Until next week, keep innovating and challenging the status quo.

See you Tuesday!

Love,

Nicole πŸ’œ

Keep Reading