Written by: a16z crypto

Translated by: Block unicorn

Introduction

Artificial Intelligence (AI) systems are disrupting the internet, which was originally designed on a human scale, by coordinating and transacting at unprecedented low costs, generating voice, video, and text that are increasingly difficult to distinguish from human activity. We have already been troubled by CAPTCHAs; now, we are beginning to see agents interact and transact like humans (as we discussed earlier).

The issue is not the existence of AI; rather, it is the internet's lack of a native method that can protect privacy while ensuring usability, and at the same time distinguish humans from machines.

This is where blockchain comes into play. Cryptography can help build better AI systems, and vice versa; this perspective may have subtle nuances. Therefore, we summarize several reasons why AI needs blockchain more than ever.

1. Increasing the Cost of AI Impersonation

AI can massively forge voices, faces, writing styles, videos, and even complete social identities: an actor can appear at an increasingly low cost in front of thousands of accounts, opinions, customers, or voters.

These impersonation strategies are not new. Any savvy scammer has always been able to hire voice actors, spoof calls, or send phishing texts. The real change lies in the cost: the cost of implementing these attacks on a large scale is decreasing.

Meanwhile, most online services assume that one account corresponds to one person. Once this assumption fails, everything downstream collapses. Detection-based methods (like CAPTCHAs) will inevitably fail because the pace of AI advancement far exceeds the tests designed to detect it.

So, what is the role of blockchain? Decentralized proof-of-human or proof-of-personhood systems make it easy for a person to become a participant, but simultaneously make it very difficult to become multiple participants. For example, scanning an iris and obtaining a world ID may be relatively easy and inexpensive, but obtaining a second world ID is nearly impossible.

This limits the supply of IDs and increases the marginal cost for attackers, making it harder for AI to achieve large-scale impersonation.

AI can forge content, but cryptography makes it more difficult to cheaply forge human uniqueness. By restoring scarcity at the identity layer, blockchain increases the marginal cost of impersonation without adding friction to normal human behavior.

2. Creating Decentralized Proof of Identity Systems

One way to prove you are human is through digital identity, which encompasses all the information a person can use to verify their identity—username, PIN, password, third-party proofs (like citizenship or credit ratings), and other credentials.

What does cryptography bring? Decentralization. Any identity system at the core of the internet becomes a point of failure. When agents transact, communicate, and coordinate on behalf of humans, whoever controls the identity effectively controls the right to participate. Issuers can revoke access, charge fees, or assist in monitoring.

Decentralization disrupts this dynamic: users, rather than platform gatekeepers, control their identities, making them more secure and resistant to censorship.

Unlike traditional identity systems, decentralized human proof mechanisms allow users to control and safeguard their identities and verify their human identity in a privacy-protecting and credibly neutral manner.

3. Creating Portable, Universal "Passports" for Agents

AI agents do not exist on a single platform. The same agent can appear in various contexts, such as chat applications, emails, phone calls, browser sessions, and APIs. However, we cannot reliably determine that interactions in these contexts point to the same agent and the same status, functions, and authorizations granted by its "owner."

Moreover, binding an agent's identity to a single platform or marketplace makes it unusable in other products and important contexts.

A blockchain-based identity layer allows agents to have portable universal "passports." These identities can include references to functions, permissions, and payment endpoints, and can be resolved from anywhere, making it harder to forge agents. This will enable developers to create more useful agents and better user experiences: agents can exist across multiple ecosystems without the worry of being locked into any specific platform.

4. Enabling Machine-Scale Payments

As AI agents increasingly represent humans in transactions, existing payment systems are gradually becoming bottlenecks. Large-scale agent payments require new infrastructure, such as micropayment systems capable of handling small transactions from multiple sources.

Many existing blockchain-based tools—such as Rollup and L2, native AI financial institutions, and financial infrastructure protocols—show potential to solve this problem, enabling near-zero-cost transactions and more granular payment splits.

Crucially, this infrastructure supports machine-scale transactions—micropayments, frequent interactions, and commercial activities between agents—that traditional financial systems cannot handle.

Micropayments can be allocated to multiple data providers, allowing users to trigger small payments to all contributors through automated smart contracts with just one interaction.

Smart contracts support executable retroactive payments triggered after a transaction is completed, thereby compensating information sources that contributed to purchasing decisions in a fully transparent and traceable manner after the transaction occurs.

Blockchain supports complex and programmable payment splits, ensuring that revenue is fairly distributed according to rules enforced by code rather than centralized decisions, thereby establishing trustless financial relationships between autonomous agents.

5. Enforcing Privacy Protection in AI Systems

At the core of many security systems lies a paradox: the more data they collect to protect users (e.g., social graphs, biometric information), the easier it becomes for AI to impersonate users.

Here, privacy and security become the same issue. The challenge is how to ensure that human proof systems default to keeping information private and obfuscate data at every step, ensuring that only humans can provide the information needed to prove their identity.

Blockchain-based systems combined with zero-knowledge proofs allow users to prove specific facts, such as PIN codes, ID numbers, or eligibility criteria (like drinking age at a bar), without revealing underlying data (e.g., the address on a driver's license).

Applications gain the necessary assurances, while AI systems are deprived of the raw data needed for imitation. Privacy is no longer an add-on feature but a core defense.

Conclusion

AI makes scaling cheap, but it is difficult to establish trust. Blockchain rebuilds trust, increases the cost of impersonation, maintains the scale of human-machine interaction, decentralizes identity, enforces privacy protection by default, and endows agents with native economic constraints.

If we want an internet where AI agents can operate without compromising trust, blockchain is not an optional infrastructure: it is key to building an AI-native internet.

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。