Original Title: 17 things we’re excited about for crypto in 2026

Translation by: Jiahua, Chaincatcher

Editor’s Note: a16z released its annual “big ideas” this week, a collaboration among various teams (Apps, American Dynamism, Bio, Crypto, Growth, Infra, Speedrun). Below are observations from a16z crypto partners and guest contributors about the future—covering themes such as agents and AI; stablecoins, tokenization, and finance; privacy and security; prediction markets, SNARKs, and other applications, as well as how we will build.

1. Higher quality and smarter stablecoin deposit/withdrawal channels

Last year, the trading volume of stablecoins was estimated to reach $46 trillion, continuously setting historical highs. To put this number into perspective: it is over 20 times the volume of PayPal; nearly 3 times that of Visa (one of the largest payment networks in the world); and is rapidly approaching the trading volume of ACH (the electronic network for financial transactions like direct deposits in the U.S.).

Today, you can send stablecoins for less than a cent and in under a second. However, the unresolved issue is how to connect these “digital dollars” to the financial rails that people use daily—in other words, the deposit/withdrawal channels for stablecoins.

A new generation of startups is filling this gap by connecting stablecoins to more familiar payment systems and local currencies. Some companies use crypto proofs to allow people to privately convert local balances into digital dollars. Others integrate regional networks, utilizing QR codes, real-time payment rails, and other features to facilitate interbank payments; while others are building truly interoperable global wallet layers and issuance platforms that allow users to spend stablecoins at everyday merchants. These approaches collectively expand the participants in the digital dollar economy and may accelerate the direct use of stablecoins as a mainstream payment method.

As these deposit/withdrawal channels mature, and digital dollars directly connect to local payment systems and merchant tools, new behaviors will emerge. Workers can receive wages in real-time across borders; merchants can accept global dollars without a bank account; applications can settle value instantly with any global user. Stablecoins will fundamentally transform from niche financial tools into the foundational settlement layer of the internet.

—— Jeremy Zhang, a16z crypto engineering team

2. Thinking about RWA tokenization and stablecoins in a more “crypto-native” way

We see strong interest from banks, fintech companies, and asset management firms in putting U.S. stocks, commodities, indices, and other traditional assets on-chain. As more traditional assets go on-chain, current tokenization is often “skeuomorphic”—rooted in the concepts of current real-world assets without leveraging crypto-native functionalities.

However, synthetic representations like perpetual futures (Perps) allow for deeper liquidity and are often easier to implement. Perps also provide easily understandable leverage, so I believe they have the strongest product-market fit (PMF) among crypto-native derivatives. I also think emerging market stocks are one of the most interesting asset classes for “perpification.” (The liquidity of “zero-day-to-expiration” or 0DTE options for certain stocks is often deeper than that of the spot market, which will be a fascinating experiment in perpification.)

It all comes down to the question of “perpification versus tokenization”; but in any case, we should see more crypto-native RWA tokenization in the coming year.

Along similar lines, in 2026, as stablecoins enter the mainstream in 2025, we will see more “native origination, not just tokenization”; the outstanding issuance of stablecoins will continue to grow.

However, stablecoins without a strong credit infrastructure look like “narrow banks,” holding specific liquid assets that are considered particularly safe. While narrow banks are an effective product, I don’t believe they will become a long-term pillar of the on-chain economy.

We have already seen many new asset management firms, curators, and protocols begin to facilitate on-chain asset-backed lending based on off-chain collateral. These loans are often issued off-chain and then tokenized. I believe the benefits of tokenization here are minimal, except perhaps for distribution to users already on-chain. That’s why debt assets should be natively issued on-chain, rather than issued off-chain and then tokenized. On-chain native issuance reduces the cost of loan servicing, back-end structural costs, and increases accessibility. The challenging part here will be compliance and standardization, but builders are already working to address these issues.

—— Guy Wuollet, a16z crypto general partner

3. Stablecoins unlock bank ledger upgrade cycles—and new payment scenarios

The software that ordinary banks run is unrecognizable to modern developers: in the 1960s and 1970s, banks were early adopters of large software systems. The second generation of core banking software began in the 1980s and 1990s (e.g., through Temenos’s GLOBUS and Infosys’s Finacle). But all of this software is aging and upgrading too slowly. Therefore, the banking industry—especially the critical core ledgers (the key databases that track deposits, collateral, and other obligations)—still often runs on mainframes, programmed in COBOL, and uses batch file interfaces instead of APIs.

Most of the world’s assets exist on these same decades-old core ledgers. While these systems are battle-tested, trusted by regulators, and deeply integrated into complex banking scenarios, they also hinder innovation. Adding key features like real-time payments (RTP) can take months, or more likely years, and requires navigating layers of technical debt and regulatory complexity.

This is where stablecoins come in. Not only have stablecoins found product-market fit and entered the mainstream over the past few years, but this year, traditional financial (TradFi) institutions have embraced them at an unprecedented level. Stablecoins, tokenized deposits, tokenized government bonds, and on-chain bonds allow banks, fintech companies, and financial institutions to build new products and serve new customers. More importantly, they can do this without forcing these organizations to rewrite their legacy systems—systems that, while aging, have been reliably running for decades. Thus, stablecoins provide institutions with a new way to innovate.

—— Sam Broner, investment partner

4. The internet as a bank

With the large-scale arrival of agents, and more business happening in the background rather than through user clicks, the way money (i.e., value!) moves needs to change.

In a world where systems act based on “intent” rather than step-by-step instructions—where AI agents identify needs, fulfill obligations, or trigger outcomes that transfer funds—value must flow as quickly and freely as today’s information. This is the role of blockchain, smart contracts, and new protocols.

Smart contracts can already settle a one-dollar payment globally in seconds. However, in 2026, emerging primitives like x402 will make that settlement programmable and responsive: agents will pay each other instantly and permissionlessly for data, GPU time, or API calls—without invoices, reconciliations, or batch processing. Developers will release software updates bundled with built-in payment rules, limits, and audit trails—without fiat integration, merchant onboarding, or banks. Prediction markets will self-settle in real-time as events unfold—odds update, agents trade, and clear globally in seconds… without custodians or exchanges.

Once value can move this way, “payment flows” will no longer be a separate operational layer but will become network behavior: banks will become part of the internet’s basic plumbing, and assets will become infrastructure. If money becomes data packets that the internet can route, then the internet is not just supporting the financial system… it becomes the financial system.

—— Christian Crowley and Pyrs Carvolth, a16z crypto listing team

5. Wealth management for everyone

Personalized wealth management services have traditionally been limited to high-net-worth clients of banks: providing customized advice and personalized portfolios across asset classes is costly and operationally complex. But as more asset classes are tokenized, strategies enabled by crypto rails—using AI recommendations and co-pilots for personalization—will be able to execute and rebalance instantly at very low costs.

This is not just about robo-advisors; everyone will have access to proactive portfolio management, not just passive management. In 2025, traditional finance increased its allocation to cryptocurrencies in portfolios (either directly or through ETPs), but this is just the beginning; in 2026, we will see platforms built for “wealth accumulation”—not just “wealth preservation”—as fintech companies (like Revolut and Robinhood) and centralized exchanges (like Coinbase) leverage their tech stack advantages to capture more market share.

At the same time, DeFi tools like Morpho Vaults will automatically allocate assets to lending markets with the best risk-adjusted returns—providing core yield distribution for portfolios. Holding remaining liquid balances as stablecoins rather than fiat, and holding them as tokenized money market funds rather than traditional money market funds, further expands yield possibilities.

Finally, retail investors will now find it easier to access more illiquid private market assets, such as private credit, pre-IPO companies, and private equity, as tokenization helps unlock these markets while still maintaining compliance and reporting requirements. As various components of balanced portfolios are tokenized (along the risk spectrum from bonds to stocks to private assets and alternative investments), they can be automatically rebalanced without cumbersome wire transfers.

—— Maggie Hsu, a16z crypto listing team

6. From “Know Your Customer” (KYC) to “Know Your Agent” (KYA)

The bottleneck of the AI agent economy is shifting from intelligence to identity.

In financial services, the number of “non-human identities” now exceeds human employees by 96 to 1—yet these identities remain unaccounted for as “ghosts.” The missing key primitive here is KYA: Know Your Agent.

Just as humans need credit scores to obtain loans, agents will need cryptographic signature credentials to transact—linking agents to their principals, their constraints, and their responsibilities. Until this exists, merchants will continue to block agents at the firewall. The industry that took decades to build KYC infrastructure now has only a few months to figure out KYA.

—— Sean Neville, co-founder of Circle and architect of USDC; CEO of Catena Labs

7. We will use AI for substantive research tasks

As a mathematical economist, it was difficult for consumer-grade AI models to understand my workflow back in January; but by November, I could give the models abstract instructions as if they were PhD students… and they sometimes returned novel and correctly executed answers. Beyond my experience, we are beginning to see AI being used more broadly in research—especially in reasoning, where models now directly assist in discoveries and even autonomously solve the Putnam problem (perhaps the hardest college-level math exam in the world).

This remains an open question: which fields will benefit most from this research assistance, and how will it assist? But I expect AI research will enable and reward a new polymath research style: one that favors the ability to speculate on the relationships between ideas and quickly infer from even more speculative answers. Those answers may not be accurate, but they can still point in the right direction (at least in some topology). Ironically, this is somewhat akin to leveraging the power of model hallucinations: when models are “smart enough,” giving them abstract space to collide may still produce nonsense—but sometimes it can unlock the door to discovery, just as people are often most creative when they do not work in linear, explicit directions.

Reasoning in this way will require a new style of AI workflow—not just agent-to-agent, but more agent-wrapping-agent—where model layers help researchers evaluate early model approaches and gradually refine the truth. I have been using this approach to write papers, while others conduct patent searches, invent new art forms, or (unfortunately) discover new types of smart contract attacks.

However, operating this research-wrapping reasoning agent ensemble will require better interoperability between models, as well as a way to identify and appropriately compensate each model's contributions—both of which cryptocurrency can help solve.

—— Scott Kominers, a16z crypto research team and Harvard Business School professor

8. The “invisible tax” on open networks

The rise of AI agents is imposing an invisible tax on open networks, fundamentally undermining their economic foundations. This disruption stems from the growing mismatch between the context layer and execution layer of the internet: currently, AI agents extract data from ad-supported websites (the context layer) to provide convenience to users while systematically bypassing the revenue streams that fund the content (such as ads and subscriptions).

To prevent the erosion of open networks (and retain the diverse content that powers AI itself), we need to deploy technological and economic solutions at scale. This could include next-generation sponsored content models, micro-ownership systems, or other new financing models. Existing AI licensing agreements have also proven to be financially unsustainable “band-aids,” often compensating content providers only a small fraction of the revenue lost due to AI siphoning traffic.

The network needs a new technological economic model where value can flow automatically. A key shift in the coming year will be from static licensing to real-time, usage-based compensation. This means testing and scaling systems—potentially leveraging blockchain to enable nano-payments and complex attribution standards—to automatically reward every entity that contributes information for the successful tasks of agents.

—— Elizabeth Harkavy, a16z crypto investment team

9. Privacy will be the most important moat in cryptocurrency

Privacy is a key feature of the world’s financial shift to on-chain. It is also a feature that nearly all existing blockchains lack. For most chains, privacy is an afterthought.

But now, privacy itself is compelling enough to distinguish one chain from all others. Privacy also does something more important: it creates a lock-in effect for the chain; if you will, a privacy network effect. Especially in a world where merely competing on performance is no longer sufficient.

Thanks to cross-chain bridge protocols, moving from one chain to another is trivial as long as everything is public. But once you make things private, it’s no longer the case: bridging tokens is easy, bridging secrets is hard. When entering and exiting private areas, there is always the risk that those monitoring the chain, memory pools, or network traffic may figure out who you are. The boundaries between private chains and public chains—even between two private chains—will leak various metadata, such as correlations of transaction times and sizes, making it easier to track someone.

Compared to many indistinguishable new chains (where fees may drop to zero due to competition—block space has become the same everywhere), blockchains with privacy can have stronger network effects. The reality is that if a “generic” chain does not have a thriving ecosystem, killer applications, or unfair distribution advantages, there is almost no reason for anyone to use it or build on it—let alone remain loyal to it.

When users are on public blockchains, they can easily transact with users on other chains—it doesn’t matter which chain they join. When users are on private blockchains, on the other hand, the chain they choose becomes much more important, as once they join, they are less likely to leave and risk exposure. This creates a winner-takes-all dynamic. And because privacy is essential for most real-world use cases, a small number of privacy chains may capture a large portion of the cryptocurrency market.

—— Ali Yahya, a16z crypto general partner

10. The (near) future of messaging is not just quantum-resistant, it is decentralized

As the world prepares for quantum computing, many encryption-based messaging applications (Apple, Signal, WhatsApp) have taken the lead and done an excellent job. The problem is that each major messaging application relies on private servers run by a single organization that we trust. These servers are easy targets for government shutdowns, backdoors, or coercion to hand over private data.

What good is quantum encryption if a country can shut down your server; if a company has the key to a private server; or even if a company owns a private server? Private servers require “trust me”—but no private server means “you don’t have to trust me.” Communication does not need a single company in the middle. Messaging needs open protocols that we do not have to trust anyone to use.

The way to achieve this is through decentralized networks: no private servers. No single application. All open-source code. Top-notch encryption—including against quantum threats. With an open network, no single person, company, nonprofit, or nation can strip us of our ability to communicate. Even if a country or company shuts down one application, 500 new versions will pop up the next day. Shut down one node, and the blockchain and other economic incentives will immediately replace it with a new node.

When people own their messages just as they own their money—that is, by owning the private keys—everything will change. Applications may come and go, but people will always maintain control over their messages and identities; end users can now own their messages, even if they do not own the application.

This is not just about being quantum-resistant and encrypted; it is about ownership and decentralization. Without these two, all we are doing is building unbreakable encryption, but it can still be shut down.

—— Shane Mac, co-founder and CEO of XMTP Labs

11. Secrets as a service

Behind every model, agent, and automation lies a simple dependency: data. But today, most data pipelines—the data that models input or output—are opaque, mutable, and un-auditable. This is fine for some consumer applications, but many industries and users (like finance and healthcare) require companies to keep sensitive data private. This is also a major barrier to institutions tokenizing real-world assets today.

So how do we enable secure, compliant, autonomous, and globally interoperable innovation while maintaining privacy? There are many ways, but I will focus on data access control: who controls sensitive data? How does it move? Who (or what) can access it?

Without data access control, anyone wanting to keep data confidential must use centralized services or build custom setups—this is not only time-consuming and expensive but also hinders the features and benefits of fully unlocking on-chain data management for traditional financial institutions. Moreover, as agent systems begin to autonomously browse, trade, and make decisions, users and institutions across industries will need cryptographic assurances, rather than “best-effort trust.”

That is why I believe we need “secrets as a service”: a new technology that provides programmable, local data access rules; client-side encryption; and decentralized key management, enforcing who can decrypt what under what conditions, and for how long… all enforced on-chain. Combined with verifiable data systems, “secrets” can then become part of the internet’s basic public infrastructure—rather than application-level patches (where privacy is often an afterthought)—thus making privacy a core infrastructure.

—— Adeniyi Abiodun, Chief Product Officer and co-founder of Mysten Labs

12. From “code is law” to “spec is law”

Recent DeFi hacks have struck well-tested protocols with strong teams, diligent audits, and years of production experience. These events highlight a disturbing reality: today’s standard security practices remain largely heuristic and case-by-case.

To mature, DeFi security needs to shift from a bug model to design-level attributes and from a “best-effort” to a “principled” approach:

• In terms of static/pre-deployment (testing, auditing, formal verification), this means systematically proving global invariants rather than verifying manually selected local invariants. AI-assisted proof tools that several teams are building can now help write specifications, propose invariants, and take on much of the manual proof engineering that has been prohibitively expensive in the past.

• In terms of dynamic/post-deployment (runtime monitoring, runtime enforcement, etc.), those invariants can become real-time “guardrails”: the last line of defense. These guardrails will be directly encoded as runtime assertions that every transaction must satisfy.

So now, we no longer assume every bug is caught, but instead enforce key security properties of the code itself, automatically rolling back any transactions that violate them.

This is not just theoretical. In practice, nearly every attack to date has triggered one of these checks during execution, potentially preventing the hacker. So the once-popular “code is law” has evolved into “spec is law”: even new types of attacks must satisfy the same security properties that maintain system integrity, making the remaining attacks either trivial or extremely difficult to execute.

—— Daejun Park, a16z crypto engineering team

13. Prediction markets becoming larger, broader, and smarter

Prediction markets have entered the mainstream, and in the coming year, as they intersect with cryptocurrency and AI, they will only become larger, broader, and smarter—while also presenting new significant challenges for builders.

First, more contracts will be listed. This means we will have access to real-time odds, not just for major elections or geopolitical events, but for a variety of nuanced outcomes and complex, cross-cutting events. As these new contracts surface more information and become part of the news ecosystem, they will raise important societal questions: about how we balance the value of this information and how to better design them to make them more transparent and auditable—which is possible with cryptocurrency.

To handle a larger volume of contracts, we need new methods to reach consensus on the truth to resolve contracts. Centralized platform resolutions are important, but controversial cases like the "Zelensky Suit Market" and the "Venezuela Election Market" show their limitations. To address these edge cases and help prediction markets expand into more useful applications, new decentralized governance and LLM oracles can help determine the truth of disputed outcomes.

AI opens up new possibilities for prediction markets—including agents automatically placing bets based on real-time data, synthesizing new contracts, and dynamically adjusting market mechanisms based on agent behavior. This will make prediction markets smarter, more responsive, and potentially unlock new use cases such as real-time risk assessment, automated hedging, and AI-driven predictions.

However, as scale increases, builders will face new challenges: ensuring market resistance to manipulation, handling the complexities of dispute resolution, and balancing information transparency with privacy. These challenges will drive innovations such as advanced cryptographic proofs and decentralized arbitration systems.

—— Andy Hall, a16z crypto research advisor and Stanford University professor of political economy

14. The Rise of "Staked Media"

The cracks in traditional media models—and their so-called objectivity—have been evident for some time. The internet has given everyone a voice, with more operators, practitioners, and builders now speaking directly to the public. Their viewpoints reflect their stakes in the world, and counterintuitively, audiences often respect them not despite these stakes, but because of them.

What’s new here is not the rise of social media, but the arrival of cryptographic tools that allow people to make publicly verifiable commitments. As AI makes generating infinite content (whether real or fictional, and capable of claiming any viewpoint or persona) cheap and easy, relying solely on what people (or bots) say feels insufficient.

Tokenized assets, programmable locks, prediction markets, and on-chain histories provide a more solid foundation of trust: commentators can make arguments and prove they are putting their money where their mouth is. Podcasters can lock tokens to prove they won’t engage in pump-and-dump schemes. Analysts can tie predictions to publicly settled markets, creating auditable records.

This is what I envision as the early form of "Staked Media": a media form that not only embraces the idea of having skin in the game but also provides proof. In this model, credibility comes not from pretending to be above it all or from baseless claims; rather, it comes from having stakes that you can make transparent and verifiable commitments to. Staked media will not replace other forms of media; it will complement what we already have. It provides a new signal: not just “trust me, I’m neutral,” but “this is the risk I’m willing to take, and here’s how you can check if I’m telling the truth.”

—— Robert Hackett, a16z crypto editorial team

15. Cryptographic Technology Provides a New Primitive Beyond Blockchain

For years, SNARKs—cryptographic proofs that allow you to verify computations without re-executing them—have primarily been a blockchain technology. The overhead has been too high: proving a computation could take 1,000,000 times more work than simply running it. This is worthwhile when distributed among thousands of validators, but impractical anywhere else.

That is about to change. By 2026, the overhead of zkVM provers will drop to about 10,000 times, with memory usage in the hundreds of megabytes—fast enough to run on mobile phones and cheap enough to be ubiquitous. Here’s a reason why 10,000 times might be a magical number: high-end GPUs have a parallel throughput about 10,000 times that of laptop CPUs. By the end of 2026, a single GPU will be able to generate proofs of CPU executions in real-time.

This could unlock a vision from old research papers: Verifiable Cloud Computing. If you have to run CPU workloads in the cloud anyway—because your computations aren’t heavy enough to require GPU acceleration, or you lack expertise, or for legacy reasons—you will be able to obtain cryptographic proofs of correctness at a reasonable overhead cost. Provers are already GPU-optimized; your code doesn’t need to be.

—— Justin Thaler, a16z crypto research team, Associate Professor of Computer Science at Georgetown University

16. Trading is Just a Waystation, Not the Endpoint of Crypto Business

It seems that today every well-run crypto company (except stablecoins and some core infrastructure) has transformed or is transforming into a trading platform. But if “every crypto company becomes a trading platform,” where does that leave everyone else? So many participants doing the same thing will erode the public's mindshare, leaving only a few big winners. This means that companies that pivot too quickly to trading miss the opportunity to build more defensible, lasting businesses.

While I have great sympathy for all those founders trying to turn their financial situations around, chasing immediate product-market fit (PMF) comes at a cost. This issue is particularly pronounced in crypto, where the unique dynamics around tokens and speculation may lead founders down the path of instant gratification in their quest for PMF… if you will, it’s a kind of “marshmallow test” (delayed gratification test).

There’s nothing wrong with trading—it’s an important market function—but it doesn’t have to be the final destination. Founders who focus on the “product” part of product-market fit may ultimately become the bigger winners.

—— Arianna Simpson, a16z crypto general partner

17. Unlocking the Full Potential of Blockchain

For the past decade, one of the biggest obstacles to building blockchain networks in the U.S. has been legal uncertainty. Securities laws have been overly broad and selectively enforced, forcing founders into a regulatory framework built for “companies” rather than “networks.” For years, mitigating legal risk has supplanted product strategy; engineers have been forced to yield to lawyers.

This dynamic has led to many strange distortions: founders being told to avoid transparency; token distributions becoming legally arbitrary; governance turning into a charade; organizational structures optimized for legal cover. Tokens have been designed to avoid economic value/no business model. Worse, noncompliant crypto projects often outpace those sincere builders.

However, regulatory clarity around crypto market structure—the likelihood of government action on this front is greater than ever—has the potential to eliminate all these distortions in the coming year. If passed, this legislation will incentivize transparency, establish clear standards, and replace “enforcement roulette” with clearer, more structured paths for financing, token issuance, and decentralization. Following the GENIUS Act, the proliferation of stablecoins has exploded; legislation around crypto market structure will be a more significant shift, but this time aimed at networks.

In other words, such regulation will enable blockchain networks to operate like networks—open, autonomous, composable, trustlessly neutral, and decentralized.

—— Miles Jennings, a16z crypto policy team and General Counsel

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。