Stablecoin payments reshape finance, AI, and the open web

Crypto rails push stablecoins into the mainstream

Across digital finance, the rise of stablecoin payments is colliding with new crypto infrastructure, AI agents, and privacy technology to reshape how value moves and how the internet itself works.

Stablecoins processed an estimated 46 trillion dollars in transaction volume last year, consistently hitting new all time highs. To benchmark that, this volume is more than 20x PayPal and nearly 3x Visa, one of the largest payment networks in the world. Moreover, stablecoin activity is rapidly approaching the volume of the ACH network for U.S. bank transfers and direct deposits.

Today, users can send a stablecoin in less than a second for less than a cent. However, connecting these digital dollars to everyday financial rails remains largely unsolved. The missing layer is robust onramps and offramps that bridge onchain value with bank accounts, local payment schemes, and merchant tools.

A new wave of startups is attacking this problem. Some leverage cryptographic proofs so users can privately swap local currency balances into digital dollars. Others plug into regional payment networks that use QR codes and real-time payment rails to enable bank-to-bank transfers. Meanwhile, more teams are building globally interoperable wallet layers and card-issuing platforms so people can spend stablecoins at everyday merchants.

Together, these approaches expand who can participate in the digital dollar economy and could accelerate stablecoins being used directly for everyday commerce. As on/off ramps mature and digital dollars integrate with local payment systems and merchant software, new behaviors will emerge. Workers may be paid in real time across borders, merchants can accept global dollars without traditional bank accounts, and apps will settle value instantly with users everywhere.

As this infrastructure scales, stablecoins will likely shift from a niche financial tool to a foundational settlement layer for the internet, underpinning everything from consumer remittances to cross-border business flows.

Rethinking tokenization and debt origination onchain

Banks, fintechs, and asset managers are showing strong interest in bringing U.S. equities, commodities, indices, and other traditional instruments onchain. However, much of this tokenization remains skeuomorphic, mirroring legacy asset structures instead of exploiting crypto-native design space.

Synthetic products such as perpetual futures offer a more native approach. Perpetuals can unlock deeper liquidity and are often simpler to implement than one-to-one asset tokens. Moreover, they provide intuitive leverage, giving them some of the strongest product-market fit among crypto derivatives.

Emerging market equities stand out as a particularly attractive asset class to “perpify.” The market for 0DTE options on certain stocks already trades with deeper liquidity than the underlying spot market, hinting at how perpification could outperform conventional tokenization. In the coming year, more crypto-native approaches to real-world asset exposure are likely to appear.

Looking ahead to 2026, stablecoins themselves are expected to evolve from simple tokenized deposits into instruments born and originated onchain. Stablecoin usage went mainstream in 2025, and outstanding issuance continues to grow. However, without a strong credit infrastructure, many current designs resemble narrow banks that only hold ultra-safe liquid assets.

While narrow banking is a valid product, it is unlikely to form the long-term backbone of an onchain economy. Instead, a new cohort of asset managers, curators, and protocols is enabling onchain asset-backed lending against offchain collateral. Today, these loans often originate offchain and are later tokenized for distribution.

Tokenization in that model offers limited benefits, aside from reaching users already onchain. That is why debt assets ultimately should be originated directly onchain rather than created offchain and wrapped later. Origination onchain can cut loan servicing and back-office structuring costs while increasing accessibility. The hard part will be compliance and standardization, yet builders are already working on those problems.

Stablecoins and the bank ledger upgrade cycle

The average bank still runs core software stacks that modern developers would barely recognize. In the 1960s and 1970s, banks were early adopters of large-scale software systems. A second generation of core banking technology arrived in the 1980s and 1990s through platforms such as Temenos GLOBUS and InfoSys Finacle. However, these systems have aged and are being upgraded slowly.

As a result, critical core ledgers that track deposits, collateral, and other obligations frequently still run on mainframes using COBOL, with batch file interfaces instead of APIs. Most global financial assets sit on these decades-old ledgers. While resilient and trusted by regulators, they also hold back innovation. For example, adding real-time payments functionality can take months or years and requires navigating layers of technical debt and regulatory complexity.

Here, stablecoins and related instruments provide a crucial bridge. Over the last few years, stablecoins have found clear product-market fit and moved firmly into the mainstream, while traditional financial institutions have embraced them at a new level. Tokenized deposits, tokenized treasuries, and onchain bonds now allow banks and fintechs to build new products without rewriting their legacy cores.

More importantly, these instruments let institutions experiment with programmable value transfer while keeping battle-tested core systems intact. Stablecoins, in effect, unlock a new cycle of bank ledger upgrades, letting innovation occur at the network edge rather than at the heart of aging stacks.

When the internet becomes the financial system

As AI agents arrive at scale and more commerce moves into the background, the way value moves must adapt. In an intent-driven architecture, systems act because an intelligent agent recognizes a need, fulfills an obligation, or triggers an outcome; users will not click through every payment step.

In that world, value must travel as fast and freely as information does today. This is precisely where blockchains, smart contracts, and new protocols fit. A smart contract can already settle a dollar payment globally in seconds. However, new primitives emerging in 2026 will make such settlement programmable and reactive.

Agents will pay each other instantly for data, GPU time, or API calls, without invoices or reconciliation. Developers will ship software updates that come bundled with payment rules, spending limits, and audit trails, all enforced without traditional merchant onboarding or bank integrations. Furthermore, prediction markets will self-settle in real time as events unfold, with odds updating, agents trading, and payouts clearing globally in seconds without custodians.

Once value flows this way, the “payment flow” ceases to be a separate operational layer and instead becomes a native network behavior. Banks turn into part of the internet’s basic plumbing; assets become infrastructure. If money becomes just another packet the internet can route, then the network does not only support the financial system — it effectively becomes the financial system.

Within this context, stablecoin payments form the connective tissue between AI agents, human users, and traditional institutions, enabling machine-to-machine commerce as naturally as email today.

Wealth management and portfolio automation for everyone

Historically, personalized wealth management has been reserved for high-net-worth clients at banks because delivering tailored advice across asset classes is expensive and operationally complex. However, as more assets become tokenized, crypto rails enable strategies that can be executed and rebalanced instantly at minimal cost.

This future is more than generic robo-advisors. Instead, everyone could access active portfolio management powered by AI recommendations and co-pilots. In 2025, traditional institutions increased portfolio exposure to crypto, with banks recommending 2-5% allocations either directly or through ETPs. That said, this shift is just beginning.

In 2026, new platforms focused on “wealth accumulation” rather than merely “wealth preservation” will emerge. Fintechs and centralized exchanges, leveraging their technical advantages, will attempt to own more of the retail market for sophisticated strategies. Meanwhile, DeFi tools such as Morpho Vaults automatically channel assets into lending markets with the best risk-adjusted yield, providing a programmable yield-bearing core allocation.

Keeping remaining liquid balances in stablecoins instead of fiat and in tokenized money market funds instead of traditional products broadens the yield universe. Moreover, retail investors now enjoy easier access to illiquid private market assets such as private credit, pre-IPO companies, and private equity, as tokenization unlocks these markets while retaining compliance and reporting.

As the components of a balanced portfolio — from bonds to equities to private and alternative assets — move onchain, they can be rebalanced automatically without wire transfers or manual processes. The result is near-institutional wealth management capabilities for a much broader base of users.

From KYC to KYA in the agent economy

In financial services, non-human identities now vastly outnumber human employees, by an estimated ratio of 96-to-1. Yet these software agents remain effectively unbanked ghosts, even as they increasingly act in markets and business workflows. The bottleneck is shifting from intelligence to identity.

The missing primitive is “Know Your Agent” or KYA. Just as humans need credit scores and verified identities to access financial services, agents will require cryptographically signed credentials that bind them to a principal, define their constraints, and clarify liability. Until that exists, many merchants and platforms will keep blocking autonomous agents at the firewall.

The same industry that spent decades building KYC infrastructure now has only months to adapt those frameworks for agents. Moreover, the intersection of AI, crypto, and programmable identity will determine how quickly agent-driven commerce can scale safely.

AI research workflows and the invisible tax on the open web

At the start of this year, some researchers struggled to get consumer AI models to understand complex workflows. By November, they could hand models abstract instructions similar to those given to doctoral students — and sometimes receive novel, correctly executed answers.

These capabilities are beginning to transform research, especially in reasoning-heavy domains. Models now assist directly in discovery and can autonomously solve challenging benchmarks such as Putnam problems, often cited as among the hardest university-level math exams. However, which fields benefit most, and how, remains an open question.

AI tools seem poised to reward a new polymath research style, where the ability to conjecture relationships between ideas and to extrapolate quickly from even speculative outputs becomes more valuable. The answers may not always be accurate, yet they can still point researchers toward fruitful directions, much like creative human brainstorming.

Achieving this will require new AI workflows, not merely agent-to-agent handoffs. Instead, researchers will rely on “agent-wrapping-agent” architectures, where layered models critique earlier attempts and distill signal from noise. This approach is already being used to draft academic papers, perform patent searches, invent new art forms, and, unfortunately, discover novel smart contract attack vectors.

Operating ensembles of reasoning agents will demand better interoperability between models and mechanisms to recognize and compensate each model’s contribution. Crypto primitives for attribution and payments can help solve both challenges.

Meanwhile, the rise of AI agents imposes an invisible tax on the open web. Agents strip data from ad-supported sites, providing convenience to users while bypassing the revenue streams — advertising and subscriptions — that fund the underlying content. Without a fix, this misalignment between the web’s context layer and execution layer threatens the sustainability of public information sources.

To prevent erosion of the open web, the ecosystem needs technical and economic innovations. Next-generation sponsored content, micro-attribution, and new funding models are all being explored. Existing AI licensing deals, however, often compensate content providers with only a fraction of the revenue lost to AI-driven traffic shifts and thus look unsustainable.

The key transition for the coming year is moving from static content licenses to real-time, usage-based compensation. Systems leveraging blockchain-enabled nanopayments and precise attribution standards could automatically reward every entity whose information contributes to an agent’s successful outcome.

Privacy as crypto’s defining moat

Privacy is critical for global finance to move onchain, yet most existing blockchains treat it as an afterthought. Today, privacy alone is compelling enough to differentiate a chain from the competition and, more importantly, to create powerful network effects.

On fully public networks, it is trivial to bridge assets from one chain to another. However, once activity becomes private, moving is much harder. Bridging tokens is easy; bridging secrets is not. Crossing between private and public zones — or even between two private chains — leaks metadata such as transaction timing and size correlations, making it easier to track users.

Because of this, privacy-preserving chains can develop significantly stronger lock-in than generic, high-throughput networks where blockspace is commoditized and fees race toward zero. If a general-purpose chain lacks a thriving ecosystem, a killer app, or a distribution edge, there is little reason for users or developers to stay loyal.

On public blockchains, users can transact easily across many networks, so chain choice matters less. On private blockchains, by contrast, users will hesitate to move once they have committed, because migration risks exposure. This dynamic could produce a winner-take-most structure where a handful of privacy-focused chains dominate, especially since privacy is essential for most real-world use cases.

Decentralized, quantum-resistant messaging

As the world anticipates quantum computing, major messaging apps such as Apple, Signal, and WhatsApp have pushed forward on stronger encryption. However, they all still depend on private servers run by single organizations. Those servers are attractive targets for governments seeking shutdowns, backdoors, or data access.

Quantum-resistant encryption means little if a country can simply unplug a company’s servers, or if a firm retains ultimate control over a private backend. Private servers inherently require “trust me” assurances. A different model says users should not need to trust anyone at all for secure communication.

To get there, messaging must be redesigned around decentralization: no private servers, no single app, and open-source code everywhere. Best-in-class encryption, including protections against quantum attacks, should sit atop open networks with no central choke points. Shut down one app and hundreds of compatible alternatives can spring up overnight.

Using blockchains and related incentives, shutting down a node simply motivates another to appear. When people hold their messages with cryptographic keys in the same way they hold money, everything changes. Applications may come and go, but users keep control of their messages and identities.

This is bigger than quantum resistance or encryption alone; it is about ownership and decentralization. Without both, society risks building unbreakable encryption that can still be switched off from the outside.

Secrets-as-a-service and spec-driven security

Behind every model, agent, and automation lies data. Yet most data pipelines — what flows into or out of models — are opaque, mutable, and unauditable. For some consumer use cases this is acceptable, but sectors such as finance and healthcare require strict privacy. It is also a major obstacle for institutions that want to tokenize sensitive real-world assets.

The key questions center on data access controls: Who governs sensitive data, how does it move, and which entities or agents can touch it? Without robust controls, anyone needing confidentiality must either use centralized services or build custom setups, both costly and slow. This reality prevents many traditional institutions from fully exploiting onchain data management.

As agentic systems begin browsing, transacting, and making autonomous decisions, both users and enterprises will demand cryptographic guarantees instead of best-effort trust. This is why “secrets-as-a-service” is becoming a necessary primitive. New technologies must provide programmable data access rules, client-side encryption, and decentralized key management that specifies who can decrypt what, under which conditions, and for how long.

Enforcing those rules onchain, combined with verifiable data systems, turns secrets into a fundamental part of the internet’s public infrastructure rather than an application-level patch. Privacy then becomes core infrastructure instead of an afterthought.

At the same time, recent DeFi hacks have struck protocols long considered battle-tested, with strong teams, thorough audits, and years in production. These incidents highlight that much of today’s security practice remains heuristic and case-by-case. To mature, DeFi security must move from bug lists to design-level properties.

On the static, pre-deployment side, that means systematically proving global invariants rather than only checking local conditions. AI-assisted proof tools can help write formal specifications, propose invariants, and reduce the manual overhead of proof engineering. On the dynamic, post-deployment side, these invariants can power runtime monitoring and enforcement.

In practice, key safety properties can be encoded as runtime assertions that every transaction must satisfy. Instead of assuming every bug was caught before launch, protocols can automatically revert any transaction that would violate core guarantees. Many past exploits would have triggered such checks and been halted mid-flight.

This evolution takes the older “code is law” concept and upgrades it to “spec is law”: even new attack vectors must still respect the same safety properties that keep the system intact. The attacks that remain become smaller or extremely difficult to execute.

Prediction markets, staked media, and SNARKs beyond blockchains

Prediction markets entered the mainstream and will become bigger, broader, and smarter as they converge with crypto and AI. In the coming year, far more contracts will be listed, providing real-time odds not just for elections and geopolitical events but also for granular, interlinked outcomes.

As these contracts surface information and integrate into the news ecosystem, they will raise questions about transparency, auditability, and social impact. Crypto can help by providing open ledgers and verifiable market structures, yet designers must carefully balance information value against potential downsides.

To handle a much greater volume of contracts, new mechanisms for resolving truth will be required. Centralized resolutions — deciding whether a given event occurred — are still important but have obvious limits. New decentralized governance frameworks and LLM-powered oracles can help settle disputed outcomes and support broader applications.

AI further expands the landscape. Agents trading on prediction platforms can scour the world for signals that provide a short-term edge, surfacing new mental models for understanding events. Besides acting as sophisticated political analysts, these agents may reveal deeper predictors of complex societal outcomes when their emergent strategies are studied.

Prediction markets do not replace polling; they complement it. Poll data can feed into markets, and AI can improve survey-taking experiences. Meanwhile, crypto tools can help verify that survey respondents are humans instead of bots, enhancing data quality.

In media, cracks in the traditional model of supposed objectivity have been visible for years. The internet gave operators, practitioners, and builders a direct channel to audiences, and many now speak from explicit stakes in the world. What is new is the arrival of cryptographic tools that let people make publicly verifiable commitments. As AI makes it trivial to generate content at scale from any persona, simple statements carry less weight. Tokenized assets, programmable lockups, prediction markets, and onchain histories can provide stronger trust foundations.

What is new is the arrival of cryptographic tools that let people make publicly verifiable commitments. As AI makes it trivial to generate content at scale from any persona, simple statements carry less weight. Tokenized assets, programmable lockups, prediction markets, and onchain histories can provide stronger trust foundations.

Meanwhile, cryptographic proofs known as SNARKs are poised to break out beyond blockchains. Historically, SNARK proving overhead was enormous — sometimes 1,000,000x more work than running a computation directly — making them useful only when amortized across thousands of validators.

By 2026, zkVM provers are expected to reach roughly 10,000x overhead with memory footprints in the hundreds of megabytes. That is fast enough for phones and cheap enough to run nearly everywhere. High-end GPUs already deliver around 10,000x more parallel throughput than laptop CPUs, so a single GPU should soon generate real-time proofs of CPU execution.

This shift could unlock verifiable cloud computing. For CPU workloads already running in the cloud — whether because they are not GPU-optimized or due to legacy reasons — developers will be able to obtain cryptographic proofs of correctness at reasonable cost. The proving layer is GPU-optimized; application code does not need to be.

Building beyond trading and toward clearer legal frameworks

Many successful crypto companies, outside of stablecoins and core infrastructure, have gravitated toward trading. When every firm chases the same business model, however, most end up competing for the same users, leaving only a few big winners and crowding out other ideas.

There is nothing inherently wrong with trading; it is an important market function. However, treating it as the final destination can be costly. Chasing immediate signals of product-market fit, especially in a token-driven ecosystem prone to speculation, can distract founders from building more defensible, long-term businesses.

The teams that stay focused on the “product” side of product-market fit, rather than only on financial volume, may ultimately capture more durable value. Trading can be a way station, not the last stop, for crypto ventures.

At the same time, one of the biggest barriers to building blockchain networks in the United States over the last decade has been legal uncertainty. Securities laws have been stretched and selectively enforced, forcing network builders into frameworks designed for conventional companies.

For years, mitigating legal risk often replaced product strategy. Founders were advised to avoid transparency; token distributions became arbitrary; governance sometimes turned into theater; and organizational structures were shaped mainly for legal cover. Tokens were even designed to avoid explicit economic value.

Crypto projects that took greater regulatory risks sometimes outpaced more cautious builders. However, crypto market structure legislation — which the U.S. government is now closer to passing than ever — could remove many of these distortions as soon as next year.

If enacted, such rules would encourage transparency, set clear standards, and replace “enforcement roulette” with structured paths for fundraising, token launches, and decentralization. After the GENIUS framework, stablecoin proliferation exploded. Comparable clarity for broader market structure could deliver an even bigger shift, this time for networks rather than just currencies.

In practice, that would allow blockchain networks to operate as true networks: open, autonomous, composable, credibly neutral, and decentralized. Combined with advances in AI, privacy, and programmable money, the next cycle of crypto innovation may look less like speculative trading and more like rebuilding the infrastructure of the internet and finance itself.

Conclusion

Across stablecoins, tokenization, AI, privacy, and governance, crypto is moving from experimental niche to core infrastructure. The next few years will determine whether these technologies become the internet’s settlement layer, the backbone of autonomous agents, and the trust fabric of a more programmable financial system.