a16z: AI needs blockchain, especially now.

The internet does not have a native mechanism that can distinguish between humans and machines while protecting privacy and usability.

Original text: AI needs blockchains — especially now

Author: @a16zcrypto

Compiled by: Bilibili News

This article is not intended as investment advice. Readers are advised to strictly abide by local laws and regulations.

AI systems are disrupting an internet originally designed according to "human scale":

It has reduced the cost of coordination, transactions, and the generation of voice, video, and text to unprecedented levels, and this content is becoming increasingly difficult to distinguish from real human behavior.

We've been plagued by CAPTCHA for years; now, we're starting to see AI agents interacting and trading like humans (something we've discussed before).

The problem is not the existence of AI itself, but that the internet does not have a native mechanism that can distinguish between humans and machines while protecting privacy and usability.

This is precisely where blockchain comes into play.

The idea that “cryptography can help build better AI systems, and vice versa” is complex and nuanced, so we’ve summarized a few key reasons here to explain why AI needs blockchain more than ever before.

1. Increase the cost of AI impersonation

AI can forge voices, faces, writing styles, videos, and even complete social identities on a large scale:

A single actor can impersonate thousands of accounts, opinions, customers, or voters, and at an increasingly lower cost.

This impersonation tactic is not new. In the past, scammers have also hired voice actors, forged phone numbers, or sent phishing text messages.

What has really changed is the price : today, the cost of carrying out these attacks on a large scale is extremely low.

At the same time, most online services still assume that an account equals a real person .

If this assumption fails, everything that follows will collapse.

Detection-based methods (such as CAPTCHA) are doomed to fail because AI will always evolve faster than the tests used to identify it.

So, what can blockchain do here?

Decentralized "proof-of-human" or "proof-of-personhood" systems make it easy to become a participant, but it remains consistently difficult to remain "only one" .

For example, scanning an iris and obtaining a World ID may not be difficult or costly, but obtaining a second one is almost impossible.

This makes it difficult for AI to impersonate on a large scale by limiting the supply of identities and increasing the marginal cost for attackers.

AI can forge content, but encryption technology makes it extremely difficult to "cheaply forge human uniqueness".

By restoring scarcity at the identity layer, blockchain significantly increases the cost of impersonation without increasing friction for ordinary users.

2. Construct a decentralized personality verification system

One way to prove you are human is through a digital identity, which includes everything an individual uses to verify their identity:

Username, PIN code, password, third-party endorsement (such as nationality or creditworthiness), and other credentials.

So, what has encryption technology brought about?

Decentralized.

Any identity system located at the center of the internet can become a single point of failure. When AI agents act as agents for humans in transactions, communication, and collaboration, whoever controls the identity system controls the right to participate .

Issuers can revoke access permissions, charge fees, and even facilitate surveillance.

Decentralization completely reverses this logic:

Users control their own identities, not the platform's gatekeepers.

This makes the identity system more secure and more resistant to censorship.

Unlike traditional identity systems, decentralized human verification mechanisms allow users to self-manage their identities and verify "I am human" in a privacy-preserving, trustworthy, and neutral manner.

3. Create a portable, universal "passport" for AI agents.

AI agents do not live in a single scenario.

The same agent can appear simultaneously in chat applications, emails, phone calls, browser sessions, and API interfaces.

However, there is currently no reliable way to confirm this:

Do these interactions in different scenarios originate from the same agent, and do they possess consistent states, capabilities, and permissions authorized by their "owner"?

Furthermore, if an agent's identity can only be bound to a certain platform or marketplace, it is almost impossible to use it in other products and ecosystems.

A blockchain-based identity layer can give AI agents a portable, universal "passport" .

These identities can include references to capabilities, permissions, and payment endpoints, and can be resolved anywhere, making it harder for agents to be forged.

This will also help developers build more useful proxies and better user experiences:

Agents can exist in multiple ecosystems without worrying about being locked into a single platform.

4. Implement a machine-scale payment system

As AI agents increasingly act as intermediaries in transactions on behalf of humans, existing payment systems are becoming a bottleneck.

Large-scale payments at the agent level require entirely new infrastructure, such as systems capable of handling a large number of micro-transactions.

Many blockchain-based tools—including Rollups, L2, AI-native financial institutions, and financial infrastructure protocols—have shown the potential to solve this problem, enabling near-zero-cost transactions and more granular profit distribution.

The key point is:

These payment tracks naturally support machine-scale transactions —micropayments, high-frequency interactions, and agent-to-agent business activities—which traditional financial systems cannot support.

• Nanoscale payments can be split among multiple data providers, enabling a single user interaction to pay micro-fees to all contributors via automated smart contracts.
• Smart contracts can trigger executable retroactive payments after a transaction is completed, compensating sources of information that influenced the purchasing decision, while maintaining complete transparency and traceability.
• Blockchain enables complex, programmable profit distribution by enforcing rules through code, rather than relying on centralized adjudication, thereby establishing trustless financial relationships between autonomous agents.

5. Enforcing privacy protection in AI systems

Many security systems suffer from a core paradox:

The more data collected to protect users (such as social relationships and biometrics), the easier it is for AI to impersonate users.

Here, privacy and security become the same issue.

The real challenge lies in ensuring that the personality verification system is privacy-preserving by default, obfuscating information at every stage, and ensuring that only humans can generate the information needed to "prove they are human".

Blockchain systems, combined with zero-knowledge proofs, enable users to prove specific facts—such as PIN codes, ID numbers, eligibility criteria (e.g., whether they have reached the legal drinking age)—without exposing underlying data (e.g., home address on a driver's license).

The application gains the necessary certainty, while the AI has no access to the original material used for imitation.

Privacy is no longer an add-on feature, but a core defense mechanism.

Conclusion

AI makes scale cheap, but makes trust difficult.

Blockchain rebuilds trust by increasing the cost of impersonation, maintaining human-scale interactions, decentralizing identities, enforcing privacy protections by default, and introducing native economic constraints to agents.

If we want an AI agent to function without eroding trust on the internet,

Blockchain is not an option, but the missing layer that enables the AI-native internet to function.

Disclaimer: As a blockchain information platform, the articles published on this site represent only the personal views of the authors and guests and do not reflect the position of Web3Caff. The information contained in the articles is for reference only and does not constitute any investment advice or offer. Please comply with the relevant laws and regulations of your country or region.