Why does AI urgently need encryption technology? 】 Author: a16z Compilation: Chopper, Foresight News AI system is subverting the Internet originally designed on a human scale. It has lowered the cost of collaboration and transaction to a new historical low, and the generated voice, video and text are increasingly difficult to distinguish from human behavior. Nowadays, we have long been troubled by human-machine verification, and currently, artificial intelligence agents are beginning to interact and transact like humans. The key to the problem is not the existence of artificial intelligence, but the lack of a native mechanism on the Internet that can distinguish between human and machine while protecting privacy and ensuring ease of use. And this is exactly where blockchain technology comes in handy. The concept that encryption technology can help create better artificial intelligence systems, and conversely, artificial intelligence can also empower encryption technology, has many deep logics behind it; Here, we summarize several reasons why artificial intelligence now requires blockchain more than ever before. Increasing the cost of artificial intelligence impersonation behavior, artificial intelligence can forge voice, facial features, writing styles, video content, and even create complete social personas, and can achieve large-scale operations: an intelligent agent can embody thousands of accounts, simulate different viewpoints, consumers, or voters, and the cost of this operation is continuously decreasing. This type of counterfeiting is not new: any fraudster with intentions has always been able to hire voice actors, forge phone calls, and send phishing messages. The real change lies in the cost: nowadays, the threshold for large-scale implementation of such fraud attacks has been significantly lowered. Meanwhile, the vast majority of online services default to 'one account corresponds to one real user'. When this premise is not established, all subsequent systems will collapse. The response methods based on detection, such as human-machine verification, will inevitably fail because the evolution speed of artificial intelligence is much faster than the detection technologies specifically designed for it. So how should blockchain work? A decentralized human authentication or identity verification system allows users to easily complete a single identity verification, but fundamentally eliminates the situation of one person having multiple identities. For example, scanning the iris to obtain a global identity may be simple and economical, but it is almost impossible for users to obtain a second identity. By limiting the number of identity identifiers issued and increasing the marginal cost for attackers, blockchain makes it difficult for artificial intelligence to achieve large-scale counterfeiting operations. Artificial intelligence can forge content, but encryption technology makes it impossible to forge unique human identities at an extremely low cost. Blockchain reshapes scarcity at the identity layer, raising the marginal cost of counterfeiting without adding additional resistance to normal human use. One way to build a decentralized human identity verification system to prove human identity is through digital identity identifiers, which cover all the information that people can use to verify their identity: usernames, personal identification codes, passwords, as well as third-party proofs (such as citizenship, credit qualifications) and other related credentials. What value can encryption technology add? The answer is decentralization. Any centralized identity system at the core of the Internet may become the fault point of the whole system. When artificial intelligence agents represent humans in conducting transactions, communication, and collaboration, whoever holds the right to identity verification actually holds the right to participate. Centralized publishers can revoke user permissions, charge fees, and even assist in monitoring activities at will. Decentralization has completely reversed this pattern: users, rather than platform gatekeepers, control their own identity information, making identity identification more secure and resistant to censorship. Unlike traditional identity systems, decentralized human authentication mechanisms allow users to independently control and safeguard their identity information, and complete human identity verification in a privacy preserving and absolutely neutral manner. Creating a portable universal 'digital passport' for artificial intelligence agents. Artificial intelligence agents are not dependent on a single platform: an intelligent agent can appear simultaneously in various chat applications, email conversations, phone conversations, browser sessions, and application programming interfaces. However, there is currently no reliable mechanism to confirm that the interactions in these different scenarios all come from the same AI agent with the same state, capabilities, and authorized by the 'owner'. In addition, if the identity of an artificial intelligence agent is only bound to a certain platform or market, it will not be able to be used in other products and important scenarios. This leads to a fragmented user experience for artificial intelligence agents, and the process of scene adaptation is cumbersome and inefficient. The identity layer based on blockchain can create a portable universal 'digital passport' for artificial intelligence agents. These identity identifiers can be associated with the abilities, permissions, and payment terminal information of intelligent agents, and can be verified in any scenario, greatly increasing the difficulty of counterfeiting artificial intelligence agents. This can also enable developers to create more practical artificial intelligence agents, bringing a better user experience: agents can run in multiple ecosystems without worrying about being bound to specific platforms. With the increasing frequency of artificial intelligence agents representing humans in transactions, the existing payment system has become a clear bottleneck in achieving large-scale payment transactions. Large scale intelligent agent payments require new infrastructure support, namely a micro payment system capable of processing small transactions from multiple sources. At present, many blockchain based tools, such as rolling scaling solutions, layer 2 networks, AI native financial institutions, and financial infrastructure protocols, have shown the potential to solve this problem, enabling almost zero cost transactions and more refined payment splitting. The key is that these blockchain payment infrastructure can support machine scale transactions, including micro payments, high-frequency interactions, and commercial transactions between intelligent agents, which traditional financial systems cannot handle. Ultra small payments can be split among multiple data providers, and through automated smart contracts, a single user interaction can trigger small payments to all relevant data providers; Smart contracts support traceable payments based on completed transactions that can be enforced. After the transaction is completed, compensation is provided to the entity that provided information support for the purchase decision. The entire process is completely transparent and traceable; Blockchain can achieve complex and programmable payment splitting and distribution, ensuring fair distribution of benefits through rules enforced by code, rather than relying on decisions made by centralized institutions, allowing autonomous agents to establish trustless financial relationships. There is a paradox at the core of many security systems in protecting privacy and security in artificial intelligence systems: the more data collected to protect users, the easier it is for artificial intelligence to impersonate user identities. In this context, privacy protection and security protection have become the same issue. The challenge we face is to make the human identity verification system default to privacy protection attributes and hide sensitive information at all stages, ensuring that only real humans can provide the relevant information needed to prove their identity. The combination of blockchain based system and zero knowledge proof technology enables users to prove specific facts, such as personal identification number, ID number number, and qualification criteria, while not disclosing underlying original data (such as address on driver's license). In this way, the application party can obtain the required authentication protection, and the artificial intelligence system is also deprived of the original data required for counterfeiting. Privacy protection is no longer an additional feature, but a core defense against artificial intelligence counterfeiting. Summary: Artificial intelligence has significantly reduced the cost of large-scale operations, but has made trust difficult to establish. And blockchain technology can reshape the trust system: raising the cost of counterfeiting behavior, preserving human scale interaction patterns, achieving decentralization of identity systems, making privacy protection the default setting, and endowing artificial intelligence agents with native economic constraints. If we want to build an Internet where AI agents can operate normally and won't destroy trust, blockchain is not an option, but a key underlying technology for building an AI native Internet, which is a core missing link of the current Internet.