Quick favor: Our records indicate that you aren’t opening this email. But records can be wrong. Please click here if you’d like to remain subscribed to Fintech Takes.
{/if}
Happy Wednesday, Fintech Listeners!
Netflix and Spotify are apparently teaming up to bring a selection of Spotify’s podcasts (including one of my very favorites, The Rewatchables) to Netflix, in an effort to compete with YouTube, which has quietly become one of the most popular podcasting platforms.
Here at Fintech Takes, we don’t post full-length video podcasts to YouTube, but perhaps we too will join in at some point? I’ve always wanted to be a television star.
Welcome back to Not Fintech Investment Advice, where Simon Taylor and I do what we do best: talk about fintech startups we’re absolutely not giving investment advice on.
This week’s lineup:
EtherFi offers a crypto-collateralized credit card that lets users borrow at 4% (compounded continuously) and earn (potentially) more than that by staking their crypto assets, flipping the net interest margin model on its head.
Lunos AI is building tireless AR agents that read emails, understand context, and collect invoices like a champ.
Circuit & Chisel is sketching out a new way for AI agents to pay other agents per use, sans subscriptions.
And Figure takes home equity lending on-chain, shrinking back-office costs, and proving that real-world assets can be tokenized.
Four different bets, one theme: money and software learning to negotiate with each other.
A decade ago, a few lines in the Durbin Amendment accidentally built an industry. The rule was meant to protect small banks by letting them keep more of their interchange fees.
Instead, it handed neobanks their business model (of the $1.67 billion Chime made in 2024, approximately 75-80% of that revenue came from the interchange).
Durbin made interchange a kind of profit engine, and sponsor banks and fintechs rode the wave.
In our conversation, Simon used the Durbin Amendment as an analogy to explain what’s happening, right now, with stablecoins and the fight to offer yield.
As I wrote on Friday, banks are scrambling to close a loophole in the GENIUS Act that’s already been turned into a business model.
The law was meant to limit the competitive impact of stablecoins by prohibiting payment stablecoin issuers from offering yield. But it left the door open for distributors (like Coinbase, among others) to do it instead.
By drawing a bright line between stablecoin issuers (who can’t offer yield) and distributors (who can), regulators created an opportunity for those distributors (if they are properly compensated by their issuer partners, as Coinbase certainly is by Circle) to acquire customers by offering market or above-market yield on stablecoins held on their platforms.
The challenge, though, is that unlike the Durbin Amendment, which created additional revenue for community banks and their fintech partners, the GENIUS Act yield loophole doesn’t generate revenue. It merely allows stablecoin distributors to spend money on customer acquisition.
This is a fundamentally different dynamic. It depends on the willingness of those distributors to invest in yield as an acquisition tool. And that willingness may not always be there (there are already signs that Coinbase’s appetite is weakening).
Is the GENIUS Act’s yield loophole a new Durbin moment? TBD.
#2: Market Efficiency, Through the Blockchain
Figure (who I’ve written about before) is working to demonstrate that the blockchain’s most useful function isn’t speculation; it’s reconciliation.
It’s bringing lending, capital markets, and asset management all on-chain. And it’s doing it at a massive scale. Outside of U.S. Treasuries, home equity loans (Figure’s core product category) are now the most tokenized real-world asset in existence.
The value of bringing a home equity loan onto a blockchain isn’t in the upfront origination process, but rather in the hand-offs that happen later, between the lender and secondary market investors. If, for example, it used to cost $100 to review each loan before it was sold and now, thanks to the standardization and transparency created by Figure, it costs $10 or $1, that’s a significant cost savings.
In a well-integrated environment, verification becomes significantly easier. And faster. The more volume that runs through the system, the less sense it makes to separate front-end origination from back-end servicing and securitization. It all becomes one holistic ecosystem.
This is already how the mortgage market works, thanks to extensive public-private sector coordination. Figure is trying to create that same level of market efficiency, without government intervention, for every other asset class.
#3: The End of Human-to-Human Finance?
For decades, B2B finance software was built for control; more dashboards, more visibility, more color-coded certainty.
Now the tools don’t just show what’s happening. They decide what to do about it.
That’s the premise behind startup Lunos AI, an AI partner that works like a human AR clerk (but 24/7 and never forgets).
It plugs into email, invoices, and ERP systems, learns how customers behave, and decides how to ask for money: when to nudge, when to negotiate, when to back off.
In Simon’s words, the Office of the CFO has been the gift that keeps on giving for fintech. But the AI transformation isn’t something where you start with the UI and work backwards; you need to start AI-native.
Anything that touches spend (like Brex, Ramp, and Arc) is already optimized (Ramp announced an AI agent for AP days after Simon and I recorded the podcast!). But on the AR side, there’s still room.
Eventually, the two worlds of AR and AP will meet in the middle.
In the future, there may be a moment when a company’s AI agent tells another’s, I’ll pay you Thursday, and they settle the terms themselves (sans emails, PDFs, and meetings). Not software for humans, but software for other software.
(Side bar: When machines start making promises on our behalf, what part of the deal still belongs to people?)
Ultimately, whoever gets to define that handshake — that is how two agents agree on money — will own B2B payments infrastructure.
This is the first episode in the new Bank Nerd Corner podcast — now under the creative direction of the one and only Kiah Haslett!
And Chris is a great first guest. In addition to being deeply knowledgable about all things banking, he also writes an excellent blog and records a really interesting podcast. I’d hate him if he wasn’t also so nice. Damnit!
I’m thrilled to be moderating this discussion at Money20/20, featuring the head of fraud at Varo and the co-founders and CEOs of SentiLink and Oscilar.
Start the AM with lending leaders unpacking the real-world journey of cash flow analytics (where to begin, how to apply it, and what it takes to make it work). Breakfast, networking, and discussion included! RSVP here.
Small panel conversation featuring Jane Barratt (Chief Advocacy Officer, MX) and yours truly (among others!), followed by drinks and hors d’oeuvres at The Grand Lux Cafe, Venetian. RSVP here.
Come for the conversation on the future of embedded finance and small business lending. Stay for the one-on-one conversations (over drinks and appetizers, of course!)
Join Salman Syed (CEO of Astrada) and yours truly for a lively AM conversation on data, AI and open banking. Who will win as the battle lines are redrawn? Bring your hunger and curiosity to Bouchon at The Venetian. RSVP here.
It’s a tough time to be a consumer. Even people making a comfortable salary are feeling the squeeze (to say nothing of those living paycheck to paycheck). Lenders are feeling the strain, too.
It’s time to consider serious changes to the processes and tools we use.
How can we detect payment stress earlier?
What communication strategies and messaging tactics are actually resonating with consumers?
What playbook changes should lenders make in the immediate future?
How can we help more consumers understand and adopt payment protection?
Join me on Nov 6 as I explore these topics (and many more) with Chris Guild (from TruStage), and Taylor Nelms (from Financial Health Network).