This article did not get the love it deserves. A single payment will soon happen using multiple different stablecoins. Onchain orchestration is what makes that possible. It is the next great frontier.

This clip is the most important section of the entire conversation. Stablecoins are accelerants to financial innovation. Global payment giants like Wise and PayPal spent over a decade and billions of dollars building global networks of banking partners, nostro and vostro accounts, and complex currency infrastructure. Today, stablecoins make it possible for a five-person startup to create a product with the same global functionality in just a few months and with almost no capital. Most people think of stablecoins simply as a better form of money, but I think they are more powerful as engines of economic opportunity and tools that give ambitious builders an edge over slower, legacy fintechs. John, Zach and Henri nailed it. Forward.

Micropayments powered by x402 is going to lead to a massive overfinancialization of content. However, it will also make the selling of literally everything more efficient.

Other governments are realizing this too….

This is how it’s done folks…. The Dollar dominates traditional FX - followed by the Euro and then Yen But if Japan moves into stables with more aggression and conviction than Europe, it could quickly close that gap.

Have been having a lot of discussions with teams of every size the last few days regarding x402 and the agentic economy. Here are the biggest takeaways that I think are very relevant (blog soon): TL;DR - Inefficient unit economics will stunt growth outside of memecoins (temporarily), interest in composability is huge but lack of resources is limiting, discovery is key and there is massive opportunities for a bunch of teams to experiment here 1. A composable agentic economy breaks with the current cost structure. It's not feasible to charge $0.01 or $0.001 for most services if an agent needs to be making hundreds or thousands of calls per "workflow" or "lifecycle". - Current default/avg. cost per action is $0.01 - Gas Fee is ~$0.000853 (per Basescan as of 1:30 PM) which is 215% increase compared to the fees paid via traditional payment processors charging ~2.7%, while on Solana I'm seeing fees of $0.001546 pulled from @PayAINetwork facilitator If an agent or agent swarm needs to consistently search, manipulate data, run sentiment analysis, store and access context, etc and does so 100 times at $0.01 that is $1. Is it possible that people will pay $1+ for hundreds of composable resources and agents to be consumed in a workflow? Yes. Is it realistic? I am not certain. Where it gets even more interesting is the facilitator. Most of them are running for free today, but some like @thirdweb and (I think) @mrdn_finance are charging fees. Assume 0.3% or 30 basis points (BPS), on $0.01 is $0.00003 which means if the gas fee is $0.00085+ the facilitator is losing $0.00082 per transaction. The cost part is interesting but I think there are arguments to be made about where teams actually make money. I.e if Coinbase makes more money from the Sequencer, then facilitator losses are irrelevant or if ThirdWeb makes money from their wallet or API products from that same dev that outpace the cost of settlement I think sure this may be a moot point BUT I don't think it's the case at scale. 2. The lack of resources I think comes from a misconception on agentic adoption. As with anything, it is easier to bring improvement to an existing customer than try to get a new one. This is why billion and trillion dollar companies spend so much time on catering to enterprises (in B2B) that are already using them. I think a specific area of the market that is almost 100% untapped in regards to x402 is agentic coding. Sure there are a couple of basic "launchpad tools" that do some things here but reality is that the moat that @cursor_ai, Codex by @OpenAI , @claudeai , @FactoryAI , etc have made around agentic coding is incredible. The problem? I can't afford to pay for all the tools. I would love to enhance every part of my workflow but it's too expensive. I cannot pay $50-100+ per month per seat/product and have dozens of products financially. The solution? x402 + invisible wallets + stablecoins + basic guardrails. The @CoinbaseDev MCP could probably work really well here as a cornerstone piece but the idea is simple (I don't claim execution here is easy). We need to bring the AI SaaS tools to x402 (community effort). Then build the pipeline to make apps like @linear make it easy to set guardrails, budgets, etc. My belief is that people will never want to deal with micro-finances. The above is also relevant because it allows us to set higher level budgets (optional granular limits) that then trickle down to the agentic workflows to write PRDs, write code, do code reviews, etc. 3. Discover is KEY. Have been thinking about this a lot and believe that kind of like how everyone is building a facilitator, the next "big" thing is going to be everyone pivoting to discovery. Sure embeddings, subgraphs, etc are all good and great but I think the key is that making agents composable is really hard and the cost of context is a limiting factory. Those building discovery tools will take an early lead who actually guide the composability in my opinion. Want to do X? Similar to how we can do vector search on embeddings; I think we will have a similar flow where discovery systems will essentially pre-sort filter based on need. Uncertain whether this will be standardized as an extension to x402 or another open protocol but the idea that an agent wants to accomplish X means that it then needs to go through a series of steps, inference, transformation, etc. The simplest solution? Discovery Agents. They have access to their own internal or public directory of resources and agents and specialize in creating agentic step-by-step plans dynamically so that general purpose agents i.e ChatGPT, Claude, etc can make a series of calls to get estimates and proposed flows and not waste context or payments based on something they don't know yet. A lot of this is my raw thoughts and I'll be refining via a blog over the next few days. If you have ideas and want to collaborate on the above just DM me!

*REMINICK: STABLECOINS WILL ELIMINATE POVERTY*

Liquidity pools suck for stablecoins. Liquidity pools are: Inefficient Pricing Tools: One dollar of USDT is not equal to one dollar of USDC, even though the dollars backing both of those tokens are EXACTLY equal to one another. This phenomenon exists because the supply and demand of individual liquidity pools are priced in, as is the need to rebalance pools after larger transactions. This pricing mechanism might work for speculative assets within DeFi, but it does not for payment-oriented stablecoins. Shallow & Narrow: In automated market maker (AMM) pools, liquidity is fragmented across countless token pairs, which means each pool has limited depth and quickly suffers from slippage on large transactions. In a world of tens of thousands of stablecoins across numerous chains, forcing stablecoins into fragmented liquidity pools makes stablecoin interoperability nearly impossible. Noncompliant: One can note that most of the traditional FX world only uses "simulated liquidity pools" rather than real liquidity pools. This is not just because of the inefficiencies outlined above, but also because liquidity pools are non-compliant with most regulatory structures and compliance obligations. For example, VASPs, EMIs, MSBs, etc. must comply with Travel Rule when moving money, which dictates that these groups know exactly who is on the other side of their liquidity. That is impossible in a traditional liquidity pool. Long live the RFQ Network.

A standard that the entire industry is going to follow....

Going to Argentina and trying to pay in the local currency.

Hot Take: There’s too much marketing in blockchain. Everyone dumps money into talking the talk, but very few can walk the walk. Once you start asking questions, you will realize very few products can actually do what they advertise.

Where do I sign up to be a stablecoin lobbyist?

Stablecoins are all I talk about, but now I get to talk about it in Amsterdam with my friends @thestablecon You should come too, so that we can all talk about stablecoins together IRL. Stablecoins

Would you rather:

OTC and MM interactions often rely on deep personal relationships, manual communications and preferred prices. This is a:

1:1 Fiat <--> Crypto is a huge step forward. But its just a step. The next step? Moving between stablecoins and chains for 1:1. FYI - being able to move between stablecoins and chains actually makes the fiat <--> stable orchestration even more efficient.

It's amazing to see remittance giants do the math and realize that they can not possibly compete with the efficiency of stablecoins. They are forced to make a choice - adopt stablecoins or die.

The stablecoin playbook has been written and is now in the early stages of execution. Great succinct analysis of the Western Union Stablecoin news from the @dynamic_xyz team.