Patrick Collision of Stripe announced the launch of a new blockchain, Tempo, saying that existing blockchains are “not optimised” for high-throughput and low-latency whereas Tempo is a ”payments-oriented L1” optimised for financial services, with a built-in token (AMM) to support neutrality with respect to different stablecoins. The “design partners” include Visa, banks and challenger banks. But they also include OpenAI and Anthropic.
Stablecoin Users
Why are these leaders in AI interested in a new blockchain for stablecoins? Well, as Michael Novogratz, the founder and chief executive officer of Galaxy Digital, recently told the Goldman Sachs Asia Leaders Conference in Hong Kong, in the near future “the biggest user of stablecoins is going to be AI”. We are seeing the birth of AI-driven economy that will be dominated by machine actors, with agents paying other agents directly for the services they require and stablecoins have now become a genuine contender to satisfy these new payment needs, a true market fit for the transfer of (non-crypto) digital assets.
A proliferation of needs, including merchant settlement, business-to-business payments, cross-border payments, retail remittances and automated payments (for example, from government) are driving that global demand for real-time, low-cost inclusive global payment solutions. I rather agree with McKinsey’s view that (and they are very polite) incumbents “may have hesitated to pursue innovative payment systems” because they do not want to disrupt existing (and very healthy) sources of revenue. Added to this natural reluctance, competition between institutions has subverted collaboration to increase net welfare.
However, it now seems that growing competition amongst serious players such as Stripe to provide stablecoin-based solutions that are nearly instant and incur predictable lower costs and (not to be underestimated in strategic terms) expand access through wallet-based rather than account-based infrastructure, means that the payments landscape is heading towards a genuine cusp, where a shift in business models will reshape the industry.
Much of the incumbent comment about the use AI in payments seems to be around fraud prevention, data analysis and process optimisations, but it must be recognised that (as to Novogratz’s point) AI will be an important user of payment systems as well as playing a role in their provision. Apart from anything else, most normal people don’t really care that much about payments and as soon as their phone, laptop, hat or badge will take care of payments for them they will happily had over responsibility to an agent.
This is why AI-powered bots will, in a srategic timeframe, start to dominate as payers and payees. This in turn means that new business models will be very different from existing ones that are built around people and organisations, applications and APIs. Out on this new frontier of autonomous agent-based systems, where agents not only respond to queries but proactively direct their own processes, orchestrate multiple tools, and execute complex tasks, agents will access the payment system in order to negotiate dynamically and to transact on behalf of individuals, businesses and other AI agents.
All of which points toward a radically different commercial landscape in which there is a natural evolution of agents into economics avatars in their own right, AI-powered bots that exchange value in return for services. Bots paying other bots are a new direction for the payments industry and they needs new rails. Now of course these new rails will not, in the longer term, be limited to stablecoins. But in the short to medium- term, stablecoins will pay a key role in enabling exploration, experimentation and (if history serves as any guide) the creation of new institutions.
(If anything Novogratz may be too conservative in his predictions, because agents will be exchanging not only stablecoins but more generalised real-world digital assets over the networks that are being built to support stablecoins right now.)
The transition to non-human users of payment systems has interesting implications on business models. For example: I remember when Collison and his brother were quoted arguing that the lack of effective payments mechanism is the reason that the web went from being an open environment and opportunity for all to an “oligopoly controlled by five companies now worth more than $3 trillion”. Ben Thompson wrote a brilliant piece about this, saying that micropayments were not viable at the dawn of the web (or since) partly because of the fee problem of the dominant retail payment system (ie, cards) but mainly, as I have previously highlighted, because micro-transactions are (as Ben phrases it) anti-human: “forcing a potential content consumer to continually decide on whether or not to pay for a piece of content is alienating, particularly when plenty of alternatives for their scarce attention exist”.
Well, indeed. But when it is my agent deciding whether to pay Forbes a dollar to read an article or to subscribe then that argument changes. You can imagine far more sophisticated interactions during which service providers would be able to communicate to agents what they wanted to make available and on what terms. taking us into a world where websites based on transactions can not only benefit from exposing themselves to agents, but in fact transact more (and potentially pay an affiliate fee).
Stablecoins Payments
I can easily imagine my shopping bot and Walmart’s supermarket bot having a negotiation around payments that is unimaginable right now. My bot will present a network token, the Walmart bot will ask my bot if it has stablecoins and will offer it extra loyalty points for using them. Then my bot will figure out whether the extra loyalty points are worth more to me that the frequent flyer miles I will get from my credit card, and so on. In the end, the cheapest solution for Walmart will be to accept stablecoins (or perhaps central bank digital currency in the longer term, but that is a discussion for another day), and with the money saved it can reward me more generously in the absence of interchange.
I am fascinated by the opportunities that agents use of stablecoins will bring, because I am sure that they are far beyond those that I can imagine right now. The next generation are already building a new financial market infrastructure on a new set of rails.