Financial Market Infrastructure Week

We somehow linked Love Island to crypto this week, the DTCC completed its first tokenization pilot, SWIFT announced a blockchain layer for tokenized bank deposits, Circle received its OCC national trust charter, and Sony Bank got preliminary approval to issue a dollar stablecoin. None of these stories are loud. None of them went viral. But together they represent the most concentrated week of financial market infrastructure news the tokenization space has had in a long time.

Market KPIs (brought to you by RWA.xyz)

📈 RWA market cap was up 1.75% WoW to $34.1 billion
🏆 Biggest RWA winner: Benji (institutional) added $225M to reach $2.6 billion
🏆 Biggest network winner: BNB Chain added $1.1B to reach $5.2 billion, driven almost entirely by iBENJI

📈 Stablecoin market cap was down slightly WoW to $299 billion
🏆 Biggest stablecoin winner: USD Go added $130M to reach $984M
🏆 Biggest network winner: Hyper EVM added $234M to reach $5.75 billion

📈 Onchain risk free rates:
Short term treasuries (1m): 3.60%
Aave / DeFi: 3.60% (onchain rates still below SOFR, holding approximately flat)

The Ostium Hack

Two days before we recorded, Ostium, the RWA-focused perps DEX on Arbitrum, was hit hard. Eighteen to twenty-four million dollars were drained from their liquidity vault, which had roughly thirty million in USDC pre-hack. The attack vector was an oracle manipulation, where the vault was tricked into believing that trades had not occurred and then paid out fake gains against those phantom positions. The oracle signer key was almost certainly compromised, though the full post-mortem has not yet been released. Funds were swapped into ETH and sent through Tornado Cash, the standard playbook.

This one hit differently for me. The Ostium team have been some of the most vocal, most cracked builders in the RWA trading space. They were OG in the offchain perps market, going back to 2023, well before Hyperliquid made the category famous. They built genuine goodwill in the community and real traction. Seeing this happen to them is genuinely awful.

That said, I've watched enough of these play out to know that recovery is possible. Euler is the canonical example: they got hacked, they found the hacker through a real investigation, and they got the funds back. The difference is often the team's ability to mobilize quickly and the nature of the attack. If anyone has information on what happened or where the funds went, the Ostium team needs it. Hit them up directly.

DTCC Completes Its First Tokenization Pilot

The DTCC announced the completion of its first tokenization pilot, converting DTCC-held securities into equivalent onchain positions. The participants included JPMorgan, Citadel Securities, Societe Generale, and DriveWealth. The pilot ran on two systems: Hyperledger Besu, which powers the DTCC's own appchain, and Canton, in a permissioned DTCC-specific environment. No public chains were involved. This is a pilot, not a product.

We covered the announcement of this pilot about six months ago. The completion matters because it confirms that the plumbing can work, at least in a controlled test environment. The DTCC has also previously signaled plans to pursue tokenization on Stellar, targeting 2027 for live deployment.

What actually changes and for whom? Honestly, for now, the beneficiaries are the broker-dealers in the DTCC network who will get cleaner back-office settlement, fewer reconciliation errors, and atomic settlement capabilities. That could be a driver toward 24/7 trading at some point, but that's a long road. For individual investors, nothing changes in the near term. The DTCC does not have a direct relationship with any individual market participant today. You're always mediated through a broker or a transfer agent. I don't see that changing.

The more interesting dynamic here is the race condition that this creates for tokenization-native broker-dealers. Alpaca currently handles something like 95% of tokenized security settlement on the introducing broker side. Oasis Pro just went live. Securitize, Figure, and others all have broker-dealers and are operating in this space. The fact that the DTCC cutover is such a high-stakes event is actually their opportunity. Think of Alpaca as the DTCC's L2. The introducing brokers are the L3s. The neobanks using those introducing brokers are the L4s. The stack is forming.

Charlie made a point I hadn't fully considered: we're romanticizing T+1 or T+0 settlement as a pure good, but T+2 might actually be a feature in some contexts. When I was at Zappo, we built a 48-hour enforced delay on vault withdrawals on purpose. We got hammered in support tickets about it. The security team's response was always calm and consistent: this is a feature. That lag time is what gives you the opportunity to catch a compromised account or a fraudulent instruction. I could actually see a world where the industry gets to atomic settlement capability but keeps some intentional friction on the high-value rails, and the T+2 window becomes a product differentiator rather than a technical limitation.

SWIFT Enters the Tokenized Deposit Layer

SWIFT announced a blockchain settlement layer for tokenized bank deposits, framed as enabling "24/7 usable bank money." The pitch is straightforward: every major bank is building its own tokenized deposit product (JPMorgan, Citi, HSBC, etc.), but none of those banks will hold each other's deposit tokens directly. Someone has to sit in the middle of interbank settlement for those positions. SWIFT already sits in that position for traditional cash flows. The move to add a blockchain layer is a natural extension of what they already do.

Charlie gave me a useful frame here: SWIFT's existing tech already resembles a rudimentary blockchain consensus mechanism. Banks submit what they think they need to settle, compare notes in sequential rounds, and flag mismatches for bilateral resolution. Multiple implementations run simultaneously, something like multi-client consensus. They've been doing turn-based distributed settlement for decades. Moving that to an actual ledger isn't as radical a departure as it sounds from the outside.

I've been asked several times in the past year who can credibly challenge SWIFT in the tokenized deposit settlement space. My honest answer is: almost no one. The compliance depth, the uptime requirements, the institutional trust, the regulatory embeddedness, these are moats that take generations to build. Ripple has been trying to displace SWIFT since 2011 or 2012. Fifteen years in, I don't think they've had a clear shot on goal. SWIFT showing up and doing this themselves was always the more probable outcome.

The one area where SWIFT might not be the right party is anywhere outside of bank-to-bank tokenized deposit settlement. Stable coin corridors, DeFi venues, high-risk payment corridors that SWIFT would never want to touch anyway, those feel like the pockets where alternatives can build. The same dynamic that allowed challenger payment companies to grow before stable coins existed (finding underserved segments that big networks painted with a broad brush as high risk) applies here. SWIFT will own the center. The edges are where it gets interesting.

Sony Bank Gets Preliminary OCC Approval for a Dollar Stablecoin

Sony Bank received preliminary OCC approval to establish Connectia Trust, a New York-based national trust bank, for the purpose of issuing a US dollar stablecoin. The stablecoin is intended to support payments across Sony's US digital ecosystem: games, anime, music, and other content. Sony Bank will capitalize the trust entity with $440 million. The stablecoin is not expected to go live until 2027.

I did not know Sony had a bank. It turns out Sony Bank is not tiny: six trillion yen in assets (roughly $39 billion), $24 billion in loans, 2.15 million accounts. Bigger than I expected.

The obvious read here is stablecoin reserves as net interest margin capture. The user buying something on the Sony PlayStation Store doesn't need to know or care whether the value in their wallet is held as a bank deposit or stablecoin reserves. They won't. But Sony will monetize the float on those reserves in a way they currently can't. That's the first-order story.

The more interesting read to me is what comes next. Sony has one of the most passionate, community-driven consumer audiences in the world. Gamers are a different breed. When I say "diehard fans," I mean people who would literally show up in cosplay to represent their favorite franchise. If you have that audience captive, and you have a bank, the incentive structures you can build get genuinely interesting. Token-gated content, stock ownership tied to in-game milestones, brokerage onboarding through the gaming experience itself, these are things that crypto has been trying to do for years but couldn't pull off because the audience for pure crypto is mercenary and the tech latency didn't work for real games.

Sony has the audience. A bank gives them the rails. Stable coins are almost always the first step on this tokenization ladder. I'm watching what comes next here with a lot of interest, even if 2027 feels far away.

The wave of gaming crypto projects from 2021 to 2023 mostly went bankrupt. I know. But those projects had the tech and no audience. This is the mirror image. We'll see if the timing is finally right.

Circle Gets Its OCC National Trust Charter

Circle received official approval for its OCC national trust charter. The stock popped 15% on the news, which Charlie found baffling, and honestly I agree. This was always going to happen. The amount of blood, sweat, and tears that went into securing this charter is enormous, and it gets distilled down to a single line item on our show notes. That's reductive, but it's also just how the scoreboard works: you either got it or you didn't.

The practical impact is meaningful. Circle can now manage its own reserves directly if it chooses. The trust charter also opens doors for other product and service lines that weren't available to them as a non-bank entity. It's a real unlock.

On the stock move: we talked a few weeks back about how we thought the market overreacted to the OpenUSD consortium announcement, given how hard distribution actually is. My read on the 15% pop here is that the market may have been partly retracing from the OpenUSD overreaction, with this news as the catalyst. OpenUSD does not have an OCC charter. Circle does. That distinction matters. Congrats to the Circle team, genuinely. This has been a long time coming.

Closing Thoughts

Four FMI stories in a single week is unusual. DTCC, SWIFT, Sony Bank, Circle. None of them are splashy. None of them generate the dopamine hit of a token pump or a viral meme. But I've been in this space long enough to remember 2015, when a wave of bank blockchain consortium projects announced themselves with great fanfare and then quietly died when the market turned. The thing that feels genuinely different now is that stable coins and tokenized deposits are decoupled from the crypto market cycle. Crypto is already in a bear market on the pure-crypto side. None of these institutional stories are slowing down because of that. The infrastructure is showing up independently of the speculation cycle. That's new. I want to see volume numbers come out of all of this, not just press releases. But the direction of travel is right.

Watch or listen to the full episode on Spotify.