a16z 2025 Outlook: AI Wallets and Stablecoin Payments Could Go Mainstream, User Experience Reigns Supreme
TechFlow Selected TechFlow Selected
a16z 2025 Outlook: AI Wallets and Stablecoin Payments Could Go Mainstream, User Experience Reigns Supreme
Multiple industries may begin tokenizing "unconventional" assets.
Navigating Web3 tides with focused insightsAuthor: a16z Crypto Research Team
Translation: TechFlow
Editor's Note: Based on insights from a16z partners across AI, U.S. dynamics, bio/health, crypto, enterprise, fintech, gaming, infrastructure, and other fields, this "Big Ideas" list highlights themes that technology developers may focus on in the coming year. Below are key points exciting the a16z crypto team. For policy and regulatory outlooks for 2025, see this article from November 2024.
AI Needs Its Own Wallet for Autonomous Behavior
As AI evolves from NPCs (non-player characters) to protagonists, it will begin acting as agents. However, until recently, AI couldn't truly participate autonomously in markets—exchanging value, revealing preferences, or coordinating resources.
We’ve already seen AI agents like @truth_terminal use cryptocurrencies for transactions, opening doors to creative content opportunities. But AI agents have even greater potential—not just fulfilling human intentions, but becoming independent network participants. As AI agent networks start managing their own crypto wallets, signing keys, and digital assets, new and interesting use cases will emerge. For example, AI could operate or validate nodes in decentralized physical infrastructure networks (DePIN), such as helping manage distributed energy. Other applications include AI agents becoming serious high-value gamers. We might even witness the first blockchain owned and operated entirely by AI.
Introducing “Decentralized Autonomous Chatbots”
Beyond AI owning wallets, another development is AI chatbots running inside Trusted Execution Environments (TEE). TEEs provide isolated environments that securely execute applications, enabling more robust designs for distributed systems. In this case, TEEs can prove that a bot operates autonomously, free from human manipulation.
Taking this further, the next big idea is a Decentralized Autonomous Chatbot (DAC). Such a chatbot could build a following by creating compelling content—whether entertaining or informative—on decentralized social media. It would earn revenue from its audience through various channels and manage its assets in cryptocurrency. The associated private keys would be managed by the TEE running the chatbot software—meaning no one except the software itself could access them.
As stakes grow, regulatory safeguards may become necessary. But the core idea is decentralization: running on a permissionless set of nodes and coordinated via consensus protocols, such a chatbot could become the first truly autonomous billion-dollar entity.
With More People Using AI, We’ll Need Unique Identity Verification
In a world rife with online impersonation, scams, multiple identities, deepfakes, and other realistic yet deceptive AI-generated content, we need “proof of personhood”—a way to confirm we’re interacting with real humans. But the new challenge isn’t fake content per se; it’s the dramatically lower cost of producing such content at scale. AI has drastically reduced the marginal cost of generating content complete with all the cues we normally rely on to judge whether something is “real.”
Therefore, now more than ever, we need ways to digitally and privately link content to individuals. “Proof of personhood” is a crucial component of digital identity. Here, it becomes a mechanism to increase the marginal cost of attacking an individual or compromising network integrity: obtaining a unique ID is free for humans but expensive and difficult for AI.
This is why privacy-preserving “uniqueness” is the next big idea in building trustworthy networks. It goes beyond merely proving personhood—it fundamentally shifts the cost structure for malicious actors. Thus, “uniqueness,” or Sybil resistance, is a non-negotiable property of any proof-of-personhood system.
From Prediction Markets to More Efficient Information Aggregation
Prediction markets gained attention in 2024 due to the U.S. election, but as an economist studying market design, I don’t expect prediction markets themselves to revolutionize 2025. Instead, they’re paving the way for tech-enabled, decentralized information aggregation mechanisms applicable across domains—from community governance and sensor networks to finance.
The past year demonstrated the concept, but note: prediction markets aren’t always ideal for aggregating information. Even for global “macro” events, they may lack reliability; for more “micro” questions, prediction pools may be too small to yield meaningful data. Yet researchers and technologists have long developed many incentive frameworks that encourage truthful knowledge sharing in diverse information settings—from data pricing and procurement mechanisms to “Bayesian truth serum” for eliciting subjective assessments—many of which are already being used in crypto projects.
Blockchains have proven ideal for implementing these mechanisms—not only because of their decentralized nature, but also because they enable open, auditable incentive schemes. Crucially, blockchains also make results publicly available, so everyone can interpret them in real time.
Enterprises Will Gradually Adopt Stablecoin Payments
Stablecoins found product-market fit over the past year—an unsurprising outcome, given they're the most economical way to send dollars globally with fast settlement. They also offer entrepreneurs an easier platform to build new payment products: no intermediaries, minimum balances, or proprietary SDKs. Yet large enterprises still haven’t realized the significant cost savings and new profit margins achievable through adoption.
While we’ve seen early interest and peer-to-peer usage, I expect 2025 to bring a wave of broader experimentation. Small and mid-sized businesses with strong brands, loyal customer bases, and high payment processing costs—like restaurants, cafes, and convenience stores—will likely lead the shift from credit cards to stablecoins. These businesses gain little from credit card fraud protection in face-to-face transactions and are most vulnerable to transaction fees (a $0.30 fee per coffee adds up!).
We should also expect larger enterprises to follow. If stablecoins accelerate banking innovation, companies will experiment with disintermediating payment processors, directly boosting their bottom line by 2%. Enterprises will also begin seeking new solutions for issues currently handled by credit card companies, such as fraud protection and identity verification.
Nations Are Exploring On-Chain Government Bonds
Tokenizing government bonds would create government-backed, interest-bearing digital assets—without the surveillance concerns tied to central bank digital currencies (CBDCs). These instruments could serve as new sources of collateral for lending and derivatives protocols in decentralized finance (DeFi), further strengthening the integrity and stability of these ecosystems.
As governments supportive of innovation explore the benefits and efficiencies of public, permissionless, and immutable blockchains, some may pilot on-chain government bonds this year. For example, the UK has explored digital securities through its financial regulator FCA’s sandbox program, and its Treasury has expressed intent to issue digital bonds.
In the U.S., with the SEC expected to require national debt clearing through traditional, cumbersome, and costly infrastructure next year, there will likely be increased discussion around how blockchains could improve bond trading transparency, efficiency, and accessibility.
We’ll See Widespread Use of “DUNA” in U.S. Blockchain Networks
In 2024, Wyoming passed a new law formally recognizing decentralized autonomous organizations (DAOs) as legal entities. DUNA, or “Decentralized Unincorporated Nonprofit Association,” was specifically designed to support decentralized governance of blockchain networks and is currently the only viable legal structure for U.S.-based projects. By adopting DUNA as a decentralized legal entity framework, crypto projects and other decentralized communities can grant their DAOs legal standing—enabling greater economic activity while protecting token holders from liability and simplifying tax and compliance needs.
DAOs are communities that manage operations of open blockchain networks and are essential tools for ensuring those networks remain open, non-discriminatory, and resistant to rent-seeking. DUNA unlocks their potential, and several projects are already actively implementing it. With the U.S. expected to push forward and accelerate its crypto ecosystem in 2025, I anticipate DUNA becoming the standard for American projects. We also expect other states to adopt similar structures (Wyoming led the way here—they were also the first state to adopt what is now the widely used LLC structure)... especially as decentralized applications beyond crypto—such as physical infrastructure/power grids—gain traction.
Liquid Democracy Moving From Online to Offline
Amid growing dissatisfaction with existing governance and voting systems, now is an ideal time to experiment with technology-supported governance models—not just online, but in the real world. I’ve previously written about how DAOs and other decentralized communities allow us to study political institutions, behaviors, and accelerated governance experiments at scale. But what if we applied these lessons to real-world governance using blockchain?
We can use blockchains for secure, private election voting—starting with low-risk pilot programs to address cybersecurity and audit concerns. More importantly, blockchains allow us to experiment with “liquid democracy” at the local level—where people can either vote directly on issues or delegate their votes. This concept, originally proposed by Lewis Carroll (author of *Alice in Wonderland* and a voting theorist), has long been impractical at scale—until now. Recent advances in computing and connectivity, combined with blockchain technology, make new forms of representative democracy possible. Crypto projects are already applying this model, generating rich data on how these systems work—see our recent research findings—that local governments and communities can learn from.
Infrastructure Reuse Will Become the Norm
Last year, many teams experimented with innovations across the blockchain stack—developing new validator sets, consensus implementations, execution engines, programming languages, and RPC APIs. While these often improved specific functionalities, they sometimes fell short on broader or fundamental capabilities. Take domain-specific SNARK programming languages: while ideal in theory for generating efficient proofs, in practice they may lag behind general-purpose languages in compiler optimizations, developer tools, learning resources, and AI coding support—and could even produce less efficient SNARKs.
Therefore, I expect that in 2025, more teams will embrace reuse—leveraging existing blockchain infrastructure components, from consensus layers and staked capital to proof systems. This approach won’t just save developers time and effort—it will let them focus on delivering unique value in their products and services.
Infrastructure is now mature enough to support web3 products ready for prime time. As in other industries, success will go to teams that skillfully navigate complex supply chains—not those who reject “not invented here” solutions.
Crypto Companies Will Start With UX, Not Let Infrastructure Dictate UX
Despite the diversity and sophistication of blockchain infrastructure, many crypto companies don’t actively choose their stack—the infrastructure effectively chooses them, shaping the user experience (UX). This happens because technical decisions at the infrastructure layer directly determine the UX of blockchain-based products and services.
Yet I believe the industry will overcome this mindset: rather than letting technology dictate the end-user experience, we should start with the desired UX and then select appropriate technologies. In 2025, more crypto product designers will begin with their target UX and work backward to infrastructure choices. Crypto startups won’t need to lock into rigid infrastructure decisions before achieving product-market fit—they can instead focus fully on finding it.
We can abstract these choices into holistic, full-stack, plug-and-play approaches, moving beyond debates over specific Ethereum Improvement Proposals (EIPs), wallet providers, intent architectures, etc. The industry is ready: abundant programmable blockspace, mature developer tools, and chain abstraction are making crypto design accessible to more people. Most tech users don’t care what language powers their favorite app. The same shift is beginning in crypto.
“Hiding the Tech” Will Drive Web3 Killer Apps
The technical strengths of blockchains make them unique—but also hinder mainstream adoption. For creators and fans, blockchains offer new ways to connect, own, and monetize. Yet jargon (“NFTs,” “zkRollups”) and complex interfaces create barriers for those who stand to benefit most. I’ve felt this firsthand in conversations with executives in media, music, and fashion who are curious about web3.
Mass adoption of consumer tech often follows a path: starting with raw technology, then iconic companies or designers simplify complexity, leading to breakout applications. Think of email—SMTP protocol hidden behind a “send” button—or credit cards, where most users today don’t care about payment mechanics. Similarly, Spotify revolutionized music not by exposing file formats, but by offering seamless playlists. As Nassim Taleb said, “Over-engineering leads to fragility; simplicity enables scale.”
Thus, I believe our industry will embrace this philosophy in 2025: “hide the tech.” The best decentralized apps already focus on intuitive interfaces—making interactions as simple as tapping a screen or swiping a card. In 2025, we’ll see more companies designing simply and communicating clearly. Successful products won’t need explanations—they’ll just solve problems.
The Crypto Industry Finally Gets Its App Stores and Discovery Platforms
When crypto apps are blocked from centralized platforms like Apple’s App Store or Google Play, user acquisition suffers. But now, emerging app stores and marketplaces are providing distribution and discovery—without gatekeeping. For instance, Worldcoin’s World App marketplace stores identity and enables access to “mini-apps,” bringing hundreds of thousands of users to multiple apps within days. Another example is the no-fee dApp store for users of Solana phones. These examples show hardware (like phones and devices) may be a key advantage for crypto app stores—just as Apple devices were crucial to early app ecosystems.
Meanwhile, other marketplaces host thousands of dApps and web3 developer tools across major blockchain ecosystems (e.g., Alchemy), and blockchains like Ronin serve as platforms for game publishing and distribution. But this trend isn’t limited to entertainment and gaming: if a product already has distribution on certain channels (like messaging apps), shifting it on-chain remains difficult (with exceptions like Telegram/TON). The same applies to apps with strong web2 distribution. Still, we may see more such transitions in 2025.
Crypto Holders Will Become Active Users
In 2024, crypto made notable progress as a political movement, with positive responses from many policymakers and politicians. It also continued evolving as a financial movement—for example, exchange-traded products (ETPs) for Bitcoin and Ethereum expanded investor participation. By 2025, crypto should evolve further—as a computing movement. So, where will new users come from?
I believe it’s time to re-engage the currently “passive” crypto holders and turn them into active users. Today, only 5–10% of crypto holders actively use crypto. We can bring the 617 million people who already own crypto into the blockchain world—especially as infrastructure improves and transaction costs fall. This means new applications will continue to emerge for both existing and new users. Meanwhile, early applications like stablecoins, DeFi, NFTs, gaming, social, DePIN, DAOs, and prediction markets are becoming more user-friendly, as communities increasingly prioritize UX and other improvements.
Multiple Industries May Begin Tokenizing “Unconventional” Assets
As infrastructure matures across crypto and other emerging technologies—driving down costs—asset tokenization will expand across industries. This will make assets once deemed inaccessible due to high costs or lack of valuation not only liquid, but meaningfully integrated into the global economy. AI engines can also leverage this data as unique datasets.
Just as fracking unlocked previously uneconomical oil reserves, tokenizing unconventional assets could redefine income models in the digital age. Once-scifi scenarios become plausible: individuals could tokenize their biometric data and rent it out via smart contracts. Early examples already exist—for instance, decentralized science (DeSci) companies using blockchain to enhance ownership, transparency, and consent in medical data collection. We don’t yet understand all implications, but these advances will let people unlock underutilized assets in decentralized ways—without relying on governments or centralized intermediaries.
The views expressed herein are those of the individuals affiliated with AH Capital Management, L.L.C. (“a16z”) and do not necessarily reflect the views of a16z or its affiliates. Certain information contained in this content is derived from third-party sources, including portfolio companies of funds managed by a16z. While such sources are believed to be reliable, a16z has not independently verified such information and makes no representations about its accuracy, completeness, or suitability for any purpose. Additionally, this content may contain third-party advertisements; a16z has not reviewed such content and does not endorse any advertising claims.
This content is for informational purposes only and should not be construed as legal, business, investment, or tax advice. You should consult your own advisors regarding any such matters. Any securities or digital assets mentioned are for illustrative purposes only and do not constitute investment advice or an offer to provide advisory services. Furthermore, this content is not directed to any investor or potential investor and should not be relied upon in making an investment decision in any fund managed by a16z. Offers to invest in a16z funds are made solely through private placement memoranda, subscription agreements, and other relevant documents, which should be read in full. References to any investments or portfolio companies do not represent all investments made by a16z-managed funds and are not guarantees that future investments will be profitable or achieve similar outcomes. A list of investments made by funds managed by Andreessen Horowitz (excluding those where the issuer has not permitted public disclosure and undisclosed public digital asset holdings) is available at a16z.com investment list.
Charts provided are for illustrative purposes only and should not be relied upon for investment decisions. Past performance is not indicative of future results. Content is effective only as of the date indicated. Any forecasts, estimates, expectations, targets, prospects, and/or opinions expressed herein are subject to change without notice and may differ from or contradict views expressed by others. Please visit a16z.com for additional important disclosures.
Join TechFlow official community to stay tuned
Telegram:https://t.me/TechFlowDaily
X (Twitter):https://x.com/TechFlowPost
X (Twitter) EN:https://x.com/BlockFlow_News
a16z
