Andreessen Horowitz’s Crypto ARM A16Z said blockchain can direct artificial intelligence towards user ownership and open standards. June 11th Report.
The paper argued that the core context and “agent passport” should exist in a self-supported wallet, and that it should provide a large linguistic model interface that allows immediate access to preferences without the need for repeated training.
The chain-based identity layer allows agents to carry verifiable records of owner, features, and payment details across email, Slack, and other platforms. Meanwhile, human rights proof systems such as World Chain’s World ID help screen bots to grow generative models.
Calculation and Payment Rails
For computing, the report has promoted a distributed physical infrastructure (DEPIN) network that pools idle GPUs and energy, turns gaming PCs and spare data center capacity into an unauthorized market, and reduces reliance on a small number of cloud providers.
The on-chain “sync layer” would have lined up the generated codebases of AI. Developers wrapped shared standards in tokenization protocols, earned automatic rewards for patching vulnerabilities, and all connected applications inherited the fixes without the need for manual updates.
This document also highlights the importance of micropayments and envisages nanomovement in a low supply chain that divides all purchases between data sources and AI that were queried before recommending the product.
The same rail can negotiate access to the web crawler. The bot posts a deposit of the token and pays a meter fee on the website “security guard” contract, but the verified person was viewed for free.
An immutable ledger can track which passages model which models and repay compensation back to publishers.
Intellectual Property and Relationship
To manage intellectual property, the report called for a public on-chain registry that incorporates ownership metadata directly into creative assets.
Smart Contract Licensing allows brands to programmatically claim or allow derivatives and turn potential infringements into loyalty flows.
At the same time, advertising may rely on wallet-based zero-knowledge proofs. Users need to reveal minimal demographic facts and receive small payments for caution while continuing to look at relevant promotions.
Looking ahead, the paper argued that AI tutors, health aides, and dating bots should be run under user keys rather than corporate dashboards.
Censor-resistant blockchains and account extraction wallets already mask the complexity of seed phrases, but optimistic or zero-knowledge coporosessers can provide the throughput required for long-term relationships that store memory in chains.
The report concluded that by embedding incentives, sources and governance directly into the protocol layer, blockchain and machine learning can be intertwined to maintain an open internet.
