Stablecoins & Agentic Payments, Why Aren't We Paying With Them Yet?
Stablecoins & Agentic Payments, Why Aren't We Paying With Them Yet?
Open to builders, researchers, and anyone curious about payments, regardless of experience level.
Stablecoins are already a $314B market, with annual transfer volume approaching Visa's. At the same time, AI agents are getting surprisingly good at shopping: they can search for products, compare prices, and even book tickets.
But there is an interesting gap.
Most people still don't use stablecoins for everyday payments. And most AI agents still stop right before payment and ask a human to confirm.
This session is about that gap.
We'll start with stablecoins: where they actually work today, why that success hasn't translated into everyday payments, and what the GENIUS Act could mean for banks and for Korea.
Then we'll move to agentic payments. If an AI agent is going to spend money on someone's behalf, how does it prove who it is? Where should its spending rules live? And if something goes wrong and the user says, "I never bought this," who takes the loss?
Join us in Seoul for a close reading and an open discussion.
Part 1. Stablecoins
Stablecoins didn't really take off because people wanted to buy coffee with crypto. Their strongest use cases have been exchange settlement, holding dollars, and moving money across borders. So what was the actual demand behind their growth?
USDT is still the largest stablecoin, and a huge amount of its transfer activity happens on Tron, largely because it's cheap to use. Does that tell us that users care less about decentralization than the industry sometimes assumes, and more about simple things like cost and convenience?
When it comes to everyday payments, what is actually holding stablecoins back? Is it bad UX — wallets, keys, gas fees — or is the bigger problem that most people simply don't have a strong reason to pay this way?
There is also the privacy question. On a public blockchain, balances and transaction histories can be visible to anyone. Is that a serious barrier to everyday payments? And if it is, what would a system that gives users privacy while still satisfying compliance requirements look like?
What changes if banks issue the stablecoin themselves? Which problems become easier, and which new trade-offs appear?
And if Korea eventually gets a KRW stablecoin, what should be designed into it from day one? Would a public blockchain make sense, or is some kind of bank consortium network more realistic?
Part 2. Agentic Payments
A lot of the infrastructure is already appearing: x402, ACP, AP2, Visa TAP, Mastercard Agent Pay, and UCP.
Agents themselves are getting more capable too. Meta's Muse, for example, can search for and book tickets, but it still asks the user before the payment happens.
So what is actually missing before we trust an agent to pay on its own?
Identity is one obvious question. When an agent makes a purchase, whose identity should matter — the company that built the agent, the person it represents, or the individual agent instance making the request?
Then there is control.
Imagine an agent is supposed to order one bottle of supplements but orders twelve instead. Each transaction is still within the user's spending limit. Who is supposed to catch that mistake? Should the user's agent policy stop it, or should the card issuer detect something unusual during authorization?
Disputes get even stranger.
Visa CE3.0 can use signals such as IP addresses and device fingerprints to help resolve "I didn't make this purchase" claims. But if thousands of AI agents are making purchases from shared cloud infrastructure, what does an IP address or device fingerprint even represent?
And finally, there is the blockchain question.
Cards already support authorization holds and conditional capture. So what can an onchain escrow system such as ERC-8183 actually do that something like Stripe cannot? And whatever the difference is, is it useful enough to justify the extra complexity?
Closing question
Five years from now, how likely is it that Korean companies use stablecoins for things like payroll or B2B settlement?
And how likely is it that an AI agent buys something in Korea without a human having to press "Confirm" first?
Bring your own questions, answers, disagreements, and opinions.
Summary link
English version: TBA
Korean version: TBA
Format
We'll start with a short introduction and some context.
Participants will then have time to silently read the summary before the discussion begins.
After that, there will be a short 5–10 minute presentation to frame the main ideas, followed by a moderator-led roundtable discussion.
Guide
Moderator: Ryan Kim (Linkedin)
Session Lead
Stablecoin: Coby (Linkedin)
AI agent payment: Ryan Kim + Junhyeok Yoo(Linkedin)
In partnership with KODA + Decipher.
Paper / source: link
Format
Short introduction and context setting to start.
A short walkthrough of the paper and its key ideas.
Questions and discussion grounded in the original source.
Moderator-led discussion with no promotions.
Session Guides:
[DESCRIPTION COMING SOON]
About Whitepaper Reading Club
We are a community of founders, researchers, and builders across Singapore, Malaysia, San Francisco, Bangkok, New York, Lagos, Taipei, and Hong Kong. We meet in person every month to read, discuss, and pressure-test the latest blockchain papers, protocols, and technical ideas.
Learn more: Website · Summaries · Calendar
We create detailed, easy-to-understand summaries for each paper and have held 80+ sessions since June 2023, covering Account Abstraction, Parallel Chains, EIPs, the Bitcoin ecosystem, and AI x Crypto. We are ecosystem-agnostic, not for profit, and focused on projects with technical, product, or social innovation.