On October 15, two seemingly unrelated pieces of news brought the next round of bets on global finance and AI governance to the table: on one side, Stripe announced collaboration with Plasma to integrate the latter into its multi-currency card infrastructure, allowing Plasma customer account balances to be spent globally through the card; on the other side, NVIDIA pledged a total of $250 million in computing resources to the UK AI security research community over the next five years, averaging about $50 million per year, and plans to invest in or take stakes in five UK startups, with computing power expected to reach institutions like the Oxford BOLD Lab, UCL SOFAIR Lab, and Imperial College through local cloud partners. On the surface, this appears to be a corporate partnership and a series of strategic investments, but the details reveal a common underlying power: Stripe, at the level of cross-border payment, aims to connect account systems from over 30 countries and regions with a card that supports USD, BRL, EUR, MXN, COP as well as USDC and OUSD, and seeks to expand its coverage to over 100 countries by 2027; meanwhile, NVIDIA, in the realm of AI security research, quietly binds the technological direction of key laboratories to its hardware ecosystem using long-term locked-in computing commitments and early-stage investments. When the gateways to payment networks and trusted computing resources are held by a few tech giants, every foundational pipeline they invest in is not just about business expansion, but also about competing for the right to define the future rules of cross-border financial flows and AI security.
The Old Order of Cross-Border Currency Exchange Faces Challenge from Stripe's New Card
The old order of cross-border payment and currency exchange has long been controlled by banks, licensed payment institutions, and card organizations: a payment from Latin America to Europe often has to loop between issuing banks, acquiring banks, and local clearing institutions, resulting in currency conversions with each passage that add costs and compliance checks, extending the time funds "drift" along the chain, making it difficult for businesses to predict receipt timing, and unable to manage currency and crypto asset exposure within the same account system. For many companies reliant on cross-border cash flows, this isn't an abstract technical issue, but a daily liquidity anxiety—should they exchange currency in advance, when will it land, and will costs spiral out of control in the last mile? These are conditions imposed under the old order.
Stripe's multi-currency card precisely targets this vulnerable point, attempting to make "cross-border" an account-level setting rather than a process that is passively endured. It allows the same account to hold balances in currencies like USD, BRL, EUR, MXN, COP while also being able to directly value assets in crypto like USDC and OUSD. When users make payments with this card, the balance logic is settled and deducted internally by Stripe, compressing the traditional chain of multiple currency exchanges and inter-organization transfers, retaining only the truly necessary parts for user experience. Barriers scattered among various banks and card organizations in the old order are weakened to mere backend configuration options by a "naturally multi-currency" card. More importantly, this infrastructure is already in operation across over 30 countries and regions, with an official plan to expand to over 100 countries by 2027, which means it is not an incremental tool for a few markets but a declaration to the entire global payment landscape: the rules for future cross-border currency exchange can be rewritten from the level of accounts and cards, without needing to rely on old intermediary chains.
After Plasma's Integration: Crypto Balances Enter Daily Consumption
For Plasma, this integration with Stripe is not just about simply placing an API into the product, but entering the core of the card infrastructure designed specifically for USDC and OUSD as a "design partner." The result is that its previously idle USDC and OUSD balances on the account page are directly connected to Stripe's multi-currency card network already deployed in over 30 countries and regions: the same card, under a unified underlying system, can treat USD, BRL, EUR, MXN, COP as well as USDC and OUSD in the account as equivalent "sources of balance" for completing online or offline payments. The changes visible to users are very intuitive—previously, crypto balances in Plasma felt like a value "lying" within the app, now they become part of a fund pool that can be used to swipe for coffee at the mall or settle on an e-commerce platform. As for how the card's revenue sharing works, how fees are settled between Stripe and Plasma, and when the product will officially launch in various regions, the briefing did not provide any information; what can be confirmed is that this is a deep collaboration around the card infrastructure itself, rather than a simple functional switch.
A more subtle change lies in the demonstration effect. Plasma has lit a lamp for other application developers with its product: the decision of whether assets like USDC and OUSD can be seen as “money for daily consumption” is not only dependent on whether there is an account balance on-chain, but also on who controls the underlying card infrastructure that can be used across multiple countries. You can have many types of crypto assets, but only those connected to a multi-currency card network like Stripe can truly enter the context of global daily payments; conversely, if they are not on this card track, they remain fenced on the side of “investment/storage.” In this logic, the card issuer and infrastructure owner are the decisive hands that determine whether crypto assets are truly "usable."
NVIDIA's $250 Million Computing Commitment to UK Security Labs
If Stripe is rewriting the underlying parameters of how "money" flows across borders, then another piece of news from October 15 concerns who will provide the computational soil for future intelligent agents. NVIDIA pledges to provide a total of $250 million in computing resources to the UK AI security research community over the next five years, which, broken down, allocates around $50 million each year specifically for the theme of "security"—in a context where computing power is seen as scarce energy, this is a large-scale targeted infusion. According to current information, these resources are expected to be accessible to research institutions through local UK cloud partners, directly connecting to security intelligent agents research teams led by Oxford's BOLD Lab, UCL's SOFAIR Lab, and Imperial College, turning "who can access computing power" into a bargaining chip for these institutions.
Computing power is just the first layer of binding. In conjunction with this commitment, NVIDIA also plans to invest in or take stakes in five UK startups; although the names, specific amounts, and rounds of these companies have not been disclosed, the structure is already clear: on one hand, supplying computing infrastructure to the research community, and on the other, pulling early-stage entrepreneurial teams into its capital network. In emerging fields like AI security, laboratories propose paradigms and methodologies, startups turn these concepts into deployable systems and services, and NVIDIA stands at the intersection of both—serving as both a provider of computing infrastructure and a stakeholder betting on application deployment. This dual strategy of "capital + infrastructure" is not merely to sell more computing power but to strategically position itself in the chosen experimental field of the UK ahead of discussions on future AI security rules and the evolution of technological paths, embedding itself into the AI security ecosystem of the UK with both capital and infrastructure.
Payment Cards and Computing Warehouses: Tech Giants Competing for Discourse Power
Putting Stripe and NVIDIA in the same frame reveals that they are engaging in the same type of business: occupying the access to infrastructure. The former uses a multi-currency card and a card infrastructure built around USDC and OUSD to connect payment paths between USD, BRL, EUR, MXN, COP and crypto assets, initially rolling out in over 30 countries and then planning to expand to over 100 countries by 2027, allowing cross-border consumption and fund usage to be completed as much as possible on its own track. Plasma, as a design partner, connects its customers' crypto asset balances directly into the "card swiping" scenario within this card infrastructure, with what is truly mastered not just being the issuance rights of a few cards, but the underlying passageways of how cross-border currencies are used, how they are accounted for, and how they flow across different legal jurisdictions.
NVIDIA is doing a similar thing in another seemingly unrelated domain. It promises to provide a total of $250 million in computing resources to the UK AI security research community over the next five years, roughly $50 million annually, and plans to invest in or take stakes in five UK startups, with computing power expected to be supplied to Oxford’s BOLD Lab, UCL’s SOFAIR Lab, and institutions like Imperial College through local cloud partners; even though this supply path still contains elements of “uncertainty,” the core point is: whoever controls the computing warehouse for training and evaluating key models has the opportunity to speak early on security benchmarks, technological pathways, and compliance thresholds. Thus, on one side are multi-currency payment cards, and on the other side is AI security computing power, two infrastructures that appear independent but push tech companies to the forefront of rule discussions. In the future, compliance standards for cross-border payments and technical norms for AI security will likely be scrutinized at the same regulatory table, with national institutions and academic teams needing to engage in repeated negotiations with companies that control payment networks and computing resources to define "what can be done" and "how it must be done," and what Stripe and NVIDIA are competing for is precisely the starting position in this defining process.
The Next Step from 30-Country Payments to Five-Year Computing Commitment
From Stripe's already deployed multi-currency card infrastructure in over 30 countries and regions, supporting USD, BRL, EUR, MXN, COP as well as USDC and OUSD, with a medium-term goal to elevate coverage to over 100 countries by 2027, to NVIDIA's commitment of a total of $250 million in computing resources locked in for the next five years after October 15, combined with plans to invest in or take stakes in five UK startups, and supplying resources to Oxford's BOLD, UCL SOFAIR, and Imperial College through local cloud partners whose credibility is not entirely guaranteed—these two seemingly separate paths are effectively turning "networks" and "computing" into infrastructure chips that can be incrementally enhanced long-term, rather than one-off projects. According to current plans, the years leading up to 2027 will be a key window for Stripe to deepen its cross-border payment network and for NVIDIA to continue fulfilling its computing and capital commitments, while also being the timeframe for other companies to observe and possibly emulate this combined strategy; if more companies begin to invest in both payment infrastructure and AI security resources simultaneously, the cost structure and compliance pathways of cross-border payments, as well as the technical standards and review processes of AI security, may likely be re-embedded into a global rule system led by tech giants, rather than being subjected to national regulation and academic consensus. This is just an analysis based on current facts, not a conclusion that has already occurred. As of October 15, we cannot be certain whether Stripe can fully achieve coverage of over 100 countries or how NVIDIA will specifically allocate its computing power among labs or startups, but what is certain is that the significance of observing this infrastructure race has surpassed the scope of a single company or announcement: on one side, there is a need to closely track the speed and extent to which crypto assets permeate daily payments through multi-currency cards and other forms, and on the other side, we must understand the flow dynamics of AI security-related computing power and funding between academic institutions and commercial companies in the coming years; these two dimensions will determine the depth of the imprint left by Stripe and NVIDIA's round of infrastructure bets on the global financial and AI security order.
Join our community, let's discuss and become stronger together!
AiCoin exclusive Hyperliquid benefits: https://app.hyperliquid.xyz/join/AICOIN88
AiCoin exclusive Aster benefits: https://www.asterdex.com/zh-CN/referral/9C50e2
On-chain Telegram community: https://t.me/AiCoinWhaleData
On-chain community: https://www.aicoin.com/link/chat?cid=N6OVMor5g
AiCoin on-chain Twitter: https://x.com/aicoinwhaledata
免责声明：本文章仅代表作者个人观点，不代表本平台的立场和观点。本文章仅供信息分享，不构成对任何人的任何投资建议。用户与作者之间的任何争议，与本平台无关。如网页中刊载的文章或图片涉及侵权，请提供相关的权利证明和身份证明发送邮件到support@aicoin.com，本平台相关工作人员将会进行核查。
