Based on a16z's annual predictions, which crypto projects are worth paying attention to?

Source: Stacy Muur

Compiled by: Ken, ChainCatcher

In mid-December, a16z released its annual "2026 Grand Vision." Below is a focus list organized based on the main arguments in its vision.

1. Reshaping Payments, Stablecoins, and Financial Rails

Argument: Stablecoins will become the settlement layer of the internet, not just the settlement layer of cryptocurrencies.

The story of stablecoins has moved past the proof-of-concept stage. Today, their annual transaction volume has reached trillions of dollars, and the question is no longer whether stablecoins are effective, but whether they can seamlessly integrate with the real-world financial system.

a16z's argument is crucial here: stablecoins are not just currencies; they are an upgrade to the ledger. If stablecoins can coexist with existing systems, providing real-time settlement, programmability, and global coverage, banks and fintech companies will not need to rewrite decades-old legacy software. This shifts the focus of value capture from issuing stablecoins to distribution, compliance, and integration.

Who Will Win in 2026?

• Embedded payments.
• Issuers (crypto-native banks) and wallets.
• Bank-level compliance + APIs.
• Global, programmable settlement.

Focus List:

• Circle (USDC): Regulatory legitimacy + distribution + liquidity advantages. Focus areas: payment APIs, bank integration, inter-agent usage.
• m0: Zero-fee USDC issuance based on modular infrastructure. Worth noting: L2 native minting, stateless agents, browser-level stablecoin user experience.
• Ether_fi: The hottest new banking competitor in 2026; offers yields through tokenized treasuries, cashback, and seamless payments; growth connecting crypto and fiat.
• Plasma: Marketed as the first neobank built entirely around stablecoins, targeting global users (especially in emerging markets) who rely on USD-backed stablecoins for their daily financial needs.
• Stablecoin (Bridge) | Stripe integration: Infrastructure for abstracting the crypto layer for deposits and withdrawals. Focus areas: integration of regional payment rails.
• x402 ecosystem: Focus areas: advancing "serving agents" and "executing by agents" programmable payments through native HTTP-level settlement.

2. RWA: Native Issuance, Not Just Tokenization

Argument: Tokenization alone is not enough. To truly enhance efficiency, the source of funds must move on-chain.

The first wave of real-world asset applications has primarily focused on tokenizing existing off-chain tools (such as loans, government bonds, and credit products) and distributing them to cryptocurrency users. While this has improved accessibility, it has largely retained the original inefficiencies: opaque underwriting processes, high service costs, slow settlement speeds, and fragmented liquidity. In many cases, tokenization is merely a new wrapper for old processes.

a16z's core insight is that the true advantage of cryptocurrencies lies not in replicating traditional financial structures but in fundamentally reshaping the credit system from the ground up. When loans are initiated on-chain, underwriting logic becomes programmable, service costs drop significantly, and risks can be priced and monitored in real-time. It is here that cryptocurrencies cease to be merely distribution channels and begin to become financial infrastructure.

Who Will Win in 2026?

• On-chain underwriting
• Transparent risk pricing
• Compliant credit mechanisms
• Abundant liquidity (often achieved through perpetual contracts)

Focus List:

• Centrifuge: On-chain credit rails for real-world assets; focus areas: institutional trading flows and default performance.
• Blackrock (BUIDL): Tokenized money market fund on Ethereum; focus areas: the amount of funds flowing into tokenized government bonds and the adoption speed in traditional finance (TradFi).
• Maple: Institutional lending with underwriter governance; focus areas: credit limit expansion and net yield vs. default risk.
• Plume: Composable credit infrastructure for programmable debt; focus areas: customizable underwriting logic and DAO use cases.
• Pendle: Yield stripping of tokenized earnings; focus areas: RWA adoption rates in government bonds and private credit.
• Ondo: Tokenized government bonds and credit funds; focus areas: compliant USDC to RWA pipeline and L2 expansion.
• Backed: Regulated ETF and bond token packaging; focus areas: DeFi-native compliance rails.

3. The Internet Becomes Banking (Smart Agents and Payments)

Argument: As AI agents begin to trade autonomously, payments can no longer be external systems tacked onto applications. They must be internet-native: instant, programmable, and fully automated.

The key shift is from user-driven execution to intent-driven execution. Agents no longer click buttons or approve invoices; they recognize conditions, fulfill obligations, and autonomously trigger actions. In this model, traditional payment processes (invoicing, batch processing, reconciliation, settlement windows) are no longer operational details but structural bottlenecks.

Blockchain introduces a different paradigm. Smart contracts can achieve global final settlement in seconds. Emerging underlying technologies further propel this process, making value transfer more responsive and composable: agents can immediately pay other agents for data, computation, or service fees upon task completion, with rules directly embedded in code, eliminating the need for intermediaries to enforce them. Currency is no longer a separate operational layer but begins to operate like network traffic, which the internet can natively handle.

Who Will Win in 2026?

• Agent-native identity.
• Programmable payment rails.
• No "human" intervention user experience.

Focus List:

• Catena: Identity and compliance infrastructure for AI agents; establishing "Know Your Agent" (KYA) standards. Focus areas: agent onboarding, enterprise-level integration.
• Nevermined: Data marketplace infrastructure for autonomous agents; enabling permissioned payments between agents for access, computation, and services. Focus areas: executable data licensing, agent monetization flows.
• KiteAI: AI-native agents with embedded payments and real-world task completion capabilities. Focus areas: intent-based automation, real economic throughput.
• ASI: Open ASI + blockchain integration layer; cross-agent collaboration and computational monetization. Focus areas: agent settlement standards, autonomous service economy.
• EigenCloud (via EigenAI): Deterministic AI agents for decentralized strategy execution and autonomous protocol operations. Focus areas: intent-driven finance, composable economic agency.
• Fetch: Multi-agent collaboration protocol enabling decentralized computing + services. Focus areas: on-chain task execution, measurable agent GDP.
• x402 implementation: Agent settlement, protocol-level payments, inter-agent transactions.

4. Privacy as the Primary Moat

Argument: Privacy leads to user lock-in. Public blockchains commoditize users.

The core insight of a16z's privacy theory is simple: blockchain space has become interchangeable, but confidentiality has not. Performance, cost, and throughput are no longer enduring differentiators. If everything is public, users can migrate freely, liquidity can bridge instantly, and applications can compete in a zero-profit environment. Privacy breaks this symmetry.

Once users, institutions, or applications put sensitive information such as state balances, strategies, counterparties, identities, and metadata into a privacy-preserving environment, switching costs naturally arise. This creates a privacy network effect: the more activity within a private domain, the higher the value of staying within it, and the greater the risk of leaving due to the risk of boundary information leakage.

Who Will Win in 2026?

• Private execution environments
• Zero-knowledge-based data access control
• Privacy protection enabled by default, rather than as an add-on feature

Focus List:

• Aztec: Private smart contracts + ZK-native Rollup. Focus areas: developer attractiveness, private DeFi primitives.
• Nillion: Decentralized MPC for private computation; institutional data hosting use cases are breakthrough signals.
• Arcium: Confidential computing layer in the Solana stack; focus areas: performance scaling and Solana-native integration.
• Aleo: ZK cloud computing platform with native incentives; enterprise-level zkCloud usage is a north star metric.
• Walrus and Seal: Key components of the Sui stack, enabling fully on-chain data and privacy.
• Payy_link: Privacy-preserving stablecoin wallet that combines crypto privacy with utility, supporting gas-free, fully private, and optionally compliant sending/receiving of stablecoins like USDC.
• Zcash: Shielded transfers via ZK-SNARKs; the launch of Halo 2 and the expansion of programmable privacy are key.
• Monero: Default privacy L1 with ring signatures; robustness under surveillance pressure remains its core moat.

5. Security: From "Code is Law" to "Norms are Law"

Argument: Audits are not enough to solve the problem. Runtime enforcement will become standard practice.

The past two years have made one thing clear: audit failures are not due to auditors' incompetence, but because audits themselves are static, localized, and fundamentally unable to comprehensively reflect dynamic systems. Today, protocols operate in adversarial environments influenced by maximum extractable value (MEV), composability, oracle delays, and extreme situations driven by incentive mechanisms, which often only emerge post-deployment and frequently under extreme market conditions.

a16z concludes (and this is a widely accepted conclusion in the industry) that security must be elevated to a higher level: from checking whether code is correct to enforcing whether system-level invariants are absolutely inviolable. This marks a shift in security policy from "code is law" to "norms are law," where protocols formally define which rules must always hold (e.g., collateral limits, value conservation, solvency conditions, ordering constraints) and continuously enforce these properties, rather than just at deployment.

Who Will Win in 2026?

• Runtime protections
• Formal specifications
• AI-assisted verification
• Continuous monitoring

Focus List:

• OpenZeppelin: Runtime rule execution and upgrade security platform with deep protocol integration.
• Trailofbits: Advanced security research firm shifting towards real-time invariant execution through tools.
• SpearbitDAO: Research-driven audit collective launching continuous audit pipelines (as opposed to static snapshots).
• Cyfrin: Smart contract security company building educational pipelines and formalization-first tools; growing influence in the modular L2 ecosystem.
• Immunefi: Unified on-chain platform for bug bounties, audits, and AI-driven threat detection to protect crypto projects from exploits.
• Startups focused on runtime monitoring.

6. Scalable Prediction Markets

Argument: Prediction markets are evolving from niche betting venues to real-time information infrastructure for the internet.

The core shift pointed out by a16z is not merely "more markets" or higher trading volumes, but the restructuring of prediction market architecture. As blockchain space costs decrease, oracle performance improves, and user experience friction diminishes, markets are no longer sporadic events related to elections or sports but begin to serve as a continuous signal extraction layer. Everything can be listed for trading: macro data releases, protocol upgrades, regulatory votes, corporate actions, and even the probability outcomes of long-tail events.

Liquidity is dispersed across thousands of micro-markets, but price discovery improves as information is no longer constrained by polls, surveys, or centralized analysts.

Who Will Win in 2026?

• Prediction markets capable of listing everything.
• AI agents trading continuously.
• Decentralized and probabilistic resolutions.

Focus List:

• Polymarket: Leading decentralized prediction platform on Polygon; monthly trading volume exceeds $1 billion, dominating the cryptocurrency space.
• Kalshi: U.S. exchange regulated by the Commodity Futures Trading Commission; monthly trading volume exceeds $1.3 billion and has top app downloads in the legal market.
• FractionAI: Dubbed the first AI agent prediction market, where agents can compete with real-time portfolios in a transparent on-chain environment.
• Opinion: High-growth event trading exchange with nominal monthly trading volume exceeding $700 million, making breakthroughs in the crypto-native betting space.
• Myriad Markets: Decentralized event market infrastructure; trading volume exceeds $10 million, with active on-chain betting covering various niches.

7. zkVMs and Verifiable Computing

Argument: Proofs leave the blockchain and enter the cloud.

The core turning point pointed out by a16z is not a slight increase in the speed of zero-knowledge proofs, but that zkVMs are crossing a threshold where general-purpose computation is no longer a luxury unique to blockchains but a viable system primitive, thanks to a dramatic reduction in proof overhead (from about 1,000,000 times to about 10,000 times), GPU-native provers, and memory usage suitable for real production environments.

Together, they open up a mechanism where ordinary CPU workloads (cloud jobs, backend services, financial models, machine learning inference, legacy enterprise code) can be executed once and then verified anywhere, transforming trust assumptions about cloud providers, data pipelines, and off-chain execution into cryptographic guarantees rather than contractual commitments, creating a world where correctness, rather than reputation, is the default security model for distributed systems.

Focus List:

• RiscZero: zkVM with general Rust computing capabilities; has a reliable lead in cloud-compatible proof space.
• Succinct: SP1 zkVM (based on RISC-V/LLVM) suitable for arbitrary Rust/LLVM code; focuses on fast verification, light clients, and off-chain data feeds, with GPU-native proofs (e.g., proving in real-time on 16 GPUs), pre-compilation for efficiency, and recursion for on-chain/off-chain use.
• Brevis_zk: zk co-processor for querying on-chain/off-chain data; designed for generating modular proofs within applications.
• Axiom_xyz: ZK co-processor for verifiable computing on on-chain/off-chain data; supports arbitrary expressive off-chain computation (e.g., historical queries) and verifies ZK proofs on-chain.
• ZKML stack implementation: Framework for verifiable ML inference via ZK proofs/SNARKs (e.g., ZKML, DSperse, JSTprove); optimizes circuits for production ML workloads (e.g., GPT-2, neural networks) to reduce overhead (proof/verification speed improved by 5-22 times).

8. Wealth Management Moves On-Chain

Argument: Active, personalized wealth management will become commonplace.

With the native tokenization of assets (including yield cash, public equities, private credit, and illiquid alternative assets), rebalancing is no longer an event but a continuous process. This process is executed by smart contracts, responding to real-time changes in interest rates, volatility, and risk premiums, rather than quarterly meetings or advisor workflows.

AI-assisted asset allocation engines are increasingly playing a co-pilot role, transforming user constraints into executable strategies; decentralized financial infrastructure technologies such as automated vaults, tiered yield products, and permissioned risk-weighted asset pools provide the depth of mechanisms needed to implement these strategies without sacrificing compliance or capital efficiency. The result is not "smart advisory," but programmable wealth, where portfolios can self-adjust based on risk curves, rotating from tokenized government bonds to credit, from Beta to Carry, from liquidity to illiquidity, with settlement, custody, and reporting abstracted away.

What Are the Winners?

• Automated rebalancing
• Yield cash
• Tokenized private markets

Focus List:

• Veda: Modular vault infrastructure supporting real-time, strategy-based rebalancing across yield primitives; aligns well with this argument.
• Upshift: Permissioned vault platform with risk monitoring strategies and KYC access; partially aligns with compliant RWA access direction.
• Midas: Tokenized government bonds and BTC yield strategies with on-chain reserve proofs; very suitable for compliant, programmable cash flows.
• Base: Coinbase's L2 Base App integrates wallets, trading, social sources, mini-programs, and on-chain yields (like USDC APY); achieves seamless, abstracted wealth management through one-click DeFi and RWA access.
• Morpho: Non-custodial lending vault that automatically rebalances between P2P and liquidity pool markets; aligns with programmable credit allocation and real-time yield optimization.
• Infinit: AI-driven super app with agent strategies for one-click automated yield, delta-neutral positions, cross-chain bridging/swapping, and RWA optimization; achieves self-adjusting wealth through multi-agent collaboration, aligning with programmable portfolio direction.

Click to learn about job openings at ChainCatcher