Original title: AI x crypto crossvers

Original by Scott Duke Kominers, Sam Broner, Jay Drain, Guy Woollet, Elizabeth Harkavy, Carra Wu, Matt Gleason, a16z crypto

Original language: Wu says block chains

The economic structure of the Internet is changing. As the open network gradually collapses into a "prompt bar", we have to think: will AI bring about a more open Internet, or will it allow us to enter the maze of the new pay wall? And who is in control of the future Internet — large central companies, or a broad user community

THIS IS THE ENTRY POINT FOR ENCRYPTION. WE HAVE DISCUSSED THE INTERSECTION BETWEEN AI AND ENCRYPTION TECHNOLOGY MANY TIMES IN THE PAST, BUT, IN SHORT, THE BLOCK CHAIN IS A WAY OF REDESIGNING INTERNET SERVICES AND NETWORK STRUCTURES TO BUILD DECENTRALIZED, CREDIBLE AND NEUTRAL SYSTEMS THAT USERS " OWN." BY RESHAPING THE ECONOMIC INCENTIVES BEHIND TODAY ' S SYSTEM, BLOCK CHAINS PROVIDE CHECKS AND BALANCES FOR THE INCREASINGLY CENTRALIZATION OF THE AI SYSTEM, THEREBY PROMOTING A MORE OPEN AND RESILIENT INTERNET。

THE IDEA THAT ENCRYPTION TECHNOLOGY CAN HELP BUILD A BETTER AI SYSTEM, AND VICE VERSA, IS NOT NEW - BUT THERE HAS BEEN A LONG ABSENCE OF A CLEAR DEFINITION. CERTAIN CROSS-CUTTING AREAS (E.G., HOW TO VALIDATE “HUMAN IDENTITY” IN THE CONTEXT OF THE PROLIFERATION OF LOW-COST AI SYSTEMS) HAVE ATTRACTED A LARGE NUMBER OF DEVELOPERS AND USERS. BUT OTHER APPLICATIONS MAY TAKE YEARS OR EVEN DECADES TO LAND. THUS, 11 AI-ENCRYPTED APPLICATIONS WERE SHARED, AND IT IS HOPED THAT MORE DISCUSSION WILL BE OPENED FOR THE INDUSTRY ON WHAT WORKS, WHAT CHALLENGES REMAIN TO BE ADDRESSED, AND HOW THEY MAY EVOLVE IN THE FUTURE。

THESE SCENARIOS ARE BASED ON TECHNOLOGIES THAT ARE CURRENTLY BEING DEVELOPED — FROM DEALING WITH LARGE AMOUNTS OF SMALL PAYMENTS TO ENSURING HUMAN RETENTION OF TITLE IN FUTURE RELATIONSHIPS WITH AI。

INTRODUCTION OF SUSTAINABLE DATA AND CONTEXT IN AI INTERACTIONS

Scott Duke Kominers: The core of generating AI is dependent on data, but in many applications the context — i.e. the state and context of interaction — is often as important as, or even more critical than, the data themselves。

Ideally, it should be possible to remember a large amount of personalized information, including the type of project you are promoting, your communication habits, preferred programming language, etc., whether it be a smart body (agent), LLM interface, or other type of AI application. In reality, however, users often have to rebuild these contexts repeatedly - not just start a new session in the same application, such as when opening a new ChatGPT or Claude window needs to recreate the context, let alone switch between different AI systems。

CURRENTLY, THE CONTEXT OF ONE GENERATION AI APPLICATION CAN HARDLY BE MIGRATED TO ANOTHER APPLICATION。

By means of block chains, AI systems can store key context elements in the form of sustainable digital assets, enabling them to be loaded at the beginning of the session and to migrate seamlessly between different AI platforms. In addition, since "forward compatibility " and " interoperability commitments " are central features of the block chain agreement, the block chain may be the only technical path to address the problem systematically。

AN INTUITIVE APPLICATION IS AN AI-LED GAME AND MEDIA FIELD WHERE USER PREFERENCES (E.G. DIFFICULTY, KEY LAYOUT, ETC.) CAN BE SUSTAINED ACROSS GAMES AND ENVIRONMENTS. BUT WHAT IS REALLY OF GREAT VALUE IS THE KNOWLEDGE-BASED APPLICATION LANDSCAPE - AI NEEDS TO UNDERSTAND THE KNOWLEDGE SYSTEMS, LEARNING PATTERNS AND COMPETENCIES OF USERS; AND MORE SPECIALIZED APPLICATIONS, SUCH AS PROGRAMMING AIDS. ALTHOUGH SOME ENTERPRISES HAVE DEVELOPED CUSTOMIZED AI TOOLS FOR THEIR OWN OPERATIONS, THEY ARE STILL UNABLE TO EFFECTIVELY MIGRATE BETWEEN DIFFERENT AI SYSTEMS USED WITHIN THE ORGANIZATION。

Organizations are only now beginning to be truly aware of the problem, and the closest to a universal solution is a customized robot with a fixed, lasting context. However, context portability among users within the platform has gradually emerged under the chain; for example, on the Poe platform, users can rent out their own self-defined robots to other users。

IF THIS TYPE OF ACTIVITY IS MOVED TO THE CHAIN, THE AI SYSTEM WITH WHICH WE INTERACT WILL BE ABLE TO SHARE A CONTEXT LAYER OF KEY ELEMENTS OF ALL OUR DIGITAL BEHAVIOR. AI IS ABLE TO UNDERSTAND OUR PREFERENCES INSTANTANEOUSLY, THEREBY BETTER FINE-TUNING AND OPTIMIZING THE EXPERIENCE. CONVERSELY, A MECHANISM SUCH AS A CHAIN-BASED INTELLECTUAL PROPERTY REGISTRATION SYSTEM, IF ALLOWED TO PERPETUATE THE CONTEXT ON THE AI REFERENCE CHAIN, COULD TRIGGER A NEW AND IMPROVED MARKET INTERACTION AROUND THE HINTS AND INFORMATION MODULES - FOR EXAMPLE, USERS CAN DIRECTLY REALIZE THEIR PROFESSIONAL COMPETENCIES IN AN AUTHORIZED MANNER, WHILE MAINTAINING DATA AUTONOMY。

Of course, as context-sharing capacity increases, it will also generate a large number of new examples and possibilities that are not currently foreseeable

2. Universal identity system for intelligence

Sam Broner: Identity — the normative record of who and what an object is — is the underlying infrastructure that supports today's digital discovery, aggregation and payment systems. However, since the platform closes such “bottom lines” to the system, users can usually only experience an identity system in a product interface that is available. Amazon, for example, assigns commodity identifiers (e.g. ASN or FNSKU) to integrate goods in a single interface and help users complete their discovery and payment; Facebook is similar: user identity determines the content of their information flows and forms the basis for their discovery of various elements in their application, including the Marketplace list of commodities, natural content and advertising。

This pattern is about to change with the rapid evolution of AI Ages. An increasing number of enterprises are using their intelligence systems for such settings as passenger service, logistics and payment, and their platforms will no longer be the traditional "one interface application" but will be distributed over multiple channels and platforms, building up deeper contexts and performing more tasks for users. But if the identity of an intelligent body is tied to a single platform or market, it will be difficult to use in other critical environments (e.g. e-mail threads, Slack channels, or other products)。

As a result, an intelligent body needs a single, portable "digital passport". Without it, it will not be possible to identify how to pay to an intelligent body, to validate its version, to search for its capabilities, to identify who it represents and to track its credibility across applications and platforms. The identity system of an intelligent body must be accompanied by a wallet, an API registration form, a change log and a social credit certificate, so that any interface (whether e-mail, Slack or other intelligent body) can be analysed and communicated in a consistent manner。

In the absence of this shared “identity language”, each system integration requires the re-establishment of the line from zero; content is found to remain ad hoc; and users will lose key context as they switch between different channels and platforms。

WE NOW HAVE THE OPPORTUNITY TO DESIGN THE SMART INFRASTRUCTURE FROM THE FIRST PRINCIPLE. SO THE QUESTION IS: HOW CAN WE BUILD A MORE RICH AND CREDIBLE AND NEUTRAL IDENTITY THAN DNS RECORDS? INSTEAD OF RESHAPING A SINGLE PLATFORM THAT BINDS TOGETHER THE FUNCTIONS OF IDENTITY, DISCOVERY, AGGREGATION, PAYMENT, ETC., SMARTS CAN RECEIVE PAYMENTS ON THEIR OWN, OPEN THEIR LIST OF CAPABILITIES AND EXIST IN MULTIPLE ECOSYSTEMS WITHOUT FEAR OF BEING LOCKED IN A PARTICULAR PLATFORM。

THIS IS WHERE ENCRYPTION WORKS AT THE INTERSECTION OF AI — A BLOCK CHAIN NETWORK THAT PROVIDES UNLICENSED COMBINATIONS THAT ENABLE DEVELOPERS TO CREATE STRONGER INTELLIGENCE AND MORE FRIENDLY USER EXPERIENCES。

Overall, vertically integrated solutions like Facebook and Amazon now experience better users -- because one of the complexities of building good products is to ensure that all components work together naturally from the top down. However, the cost of this facility is becoming increasingly high, especially in the context of lower software costs for the construction, aggregation, promotion, commercialization and distribution of smarts, as well as the growing reach of smart applications。

Much remains to be done to achieve the user experience of the vertical integration platform, but once a credible and neutral layer of intelligence is built, entrepreneurs can truly have their own passport. This will also facilitate extensive experiments and innovations in distribution models and interactive design。

3. Proof of Humankind, PoP

Jay Drain Jr. and Scott Duke Kominers: With the spread of AI — whether robots and smarts that interact with web pages or deep-rooted manipulation of social media — it is becoming more difficult for us to judge whether the objects we interact with online are real human beings. This erosion of trust is not a concern for the future, but a reality that is taking place. The automated account numbers from the X review of the Navy to the dating software are becoming blurred. In such an environment, "human proof" is becoming a key infrastructure for the Internet。

ONE WAY TO VERIFY THAT YOU ARE HUMAN IS TO USE DIGITAL IDENTITIES, INCLUDING THE CENTRALIZED IDENTIFICATION SYSTEM USED BY INSTITUTIONS LIKE TSA. DIGITAL ID COVERS ALL INFORMATION THAT A USER CAN USE TO PROVE HIS/HER IDENTITY - USER NAME, PIN CODE, PASSWORD, AND CERTIFICATES FROM THIRD PARTIES (SUCH AS NATIONALITY, CREDIBILITY OR CREDIT STATUS). THE VALUE OF CENTRALIZATION IS CLEAR: WHEN IDENTITY DATA ARE STORED IN A CENTRALIZED SYSTEM, THE ISSUER CAN WITHDRAW ACCESS RIGHTS, CHARGE FEES AND EVEN ASSIST IN MONITORING. DECENTRIZATION, ON THE OTHER HAND, DESTABILIZES THE STRUCTURE: USERS, RATHER THAN PLATFORM GATEKEEPERS, TAKE CONTROL OF THEIR IDENTITIES, MAKE THEM SAFER AND RESISTANT TO CENSORSHIP。

Unlike traditional identity systems, de-centreding mechanisms for the proof of humanity (e.g. World's Proof of Human) allow users to manage identity data autonomously and to validate themselves as "humans " in a manner that protects privacy and is credible and neutral. A decentralized PP, similar to a driver's licence, which can be used whenever and wherever it is issued, can serve as a generic base module for reuse on any platform, including those that do not yet exist. In other words, a Pop based on a block chain has "forward compatibility" because it provides:

Portability: Agreements are open standards and any platform can be integrated. The centralization of the Pops can be managed through public infrastructure and is completely controlled by users. This means that PP is naturally portable and that any platform is or will be compatible with it。

Accessability without permission_: The Platform can choose whether or not to support a PoP identity without a central API approval that may impose discriminatory restrictions on different uses。

The central challenge in this area is the adoption rate. Currently, there are no large-scale "poP" applications in the real world, but we expect that there will be a significant acceleration in the spread of the PoP once the number of users has reached critical scale, a number of early partners have emerged, and a "killer application" that drives user demand. Each application that uses a digital ID standard increases the value of this type of ID to the user; in turn encourages more users to access the ID; and a larger user size reverses the appeal of applying an integrated ID standard to validate "human nature". (In addition, the network effect can spread rapidly because the chain ID is designed to be interoperability. I'm not sure

We have seen mainstream consumer applications such as games, friends and social media announce working with World ID to ensure that users actually interact with real human beings — even the specific objects they expect — when playing, talking or trading. At the same time, new identity agreements have emerged this year, such as the Solar Aviation Service (SAS). While SAS itself is not a PoP issuer, it allows users to associate the data under the chain (e.g., the KYC results required for compliance, investor certification, etc.) in a private way to the Solana wallet, thereby building the decentrized identity of users. These indications suggest that the turning point of the decentrization of the PP may not be far away。

Proof of Personhood means more than "stop robots." It aims to create a clear boundary between AI ' s intelligent body and human networks, allowing users and applications to distinguish between the different interactions between human beings and machines, thus creating conditions for better quality, safer and more authentic digital experience。

4. Decentralised physical infrastructure for AI (DePIN)

Guy Woollet: AI, although a digital service, is increasingly constrained by physical infrastructure. To go to the Central Physical Infrastructure Network (DEPIN) — a new world-wide architecture and operating model that promises to make the computing infrastructure that supports AI more democratic, less costly, more resilient and more review-resistant。

Why? AI The two main bottlenecks to development are energy and chip access, respectively. Decentralised energy systems can provide more electricity, and developers are using DePIN to integrate idle chips from games PC, data centres and other sources. These computing devices can together form a non-licensed calculator market, thus creating a level playing field for the construction of new AI products。

Other applications include distributed training and fine-tuning large language models (LLMs) and the construction of distributed reasoning networks (model inference). Decentreized training and reasoning can significantly reduce costs because they take advantage of otherwise inactive computing resources. At the same time, such structures have natural anti-censorship capabilities to ensure that developers are not "down " or restricted access by relying on super-large cloud service providers (hyperscalers, central cloud infrastructure providers that provide large-scale, scalable computing resources)。

THE CONCENTRATION OF THE AI MODEL IN THE HANDS OF A SMALL NUMBER OF COMPANIES HAS BECOME A LONG-STANDING CONCERN; DECENTRIZATION NETWORKS CAN HELP BUILD MORE COST-EFFECTIVE, RESISTANT AND SCALABLE AI SYSTEMS。

5. ESTABLISHMENT OF INFRASTRUCTURE AND SECURITY MECHANISMS FOR INTERFACE BETWEEN AI SMARTS, END-SERVICE PROVIDERS AND USERS

Scott Duke Kominers: AI will increasingly need to work independently with other AI without direct human control, as the AI tool grows its capacity to handle complex tasks and implement multi-level interactive chains。

FOR EXAMPLE, AN AI SMART BODY MAY NEED TO REQUEST SPECIFIC DATA FOR A GIVEN CALCULATION, OR CALL ON OTHER INTELLIGENT BODIES WITH SPECIALIZED CAPABILITIES TO PERFORM THE TASK - – FOR EXAMPLE, THE STATISTICAL ANALYSIS INTELLIGENCE IS RESPONSIBLE FOR BUILDING AND RUNNING MODEL SIMULATIONS OR FOR MOBILIZING IMAGES TO GENERATE INTELLIGENCE TO ASSIST IN THE PRODUCTION OF MARKETING MATERIAL. AI SMARTS WILL ALSO CREATE SIGNIFICANT VALUE IN THE IMPLEMENTATION OF END-TO-END SERVICES, SUCH AS THE COMPLETE REPLACEMENT OF USERS TO COMPLETE A TRANSACTION PROCESS: TO LOCATE AND BOOK FLIGHTS ACCORDING TO PREFERENCES, OR TO AUTOMATICALLY DISCOVER AND PURCHASE NEW BOOKS THAT MEET USER PREFERENCES。

AT PRESENT, THERE IS NO "ONE-SIZE-FITS-ALL" MARKET FOR SMARTS. SUCH CROSS-INTELLECTUAL BODY REQUESTS CAN NORMALLY BE PERFORMED ONLY BY VISIBLE API CALLS, OR ONLY WITHIN SOME CLOSED AI INTELLIGENT BODY ECOSYSTEMS, AS INTERNAL FUNCTIONS。

MORE BROADLY, MOST AI INTELLIGENTS NOW OPERATE IN ISOLATED ECOSYSTEMS: API IS RELATIVELY CLOSED AND LACKS UNIFORM ARCHITECTURE STANDARDS. FOR EXAMPLE, BLOCK-CHAIN TECHNOLOGY CAN HELP ESTABLISH OPEN STANDARDS FOR AGREEMENTS, WHICH IS ESSENTIAL FOR SHORT-TERM ADOPTION; IN THE LONG TERM, IT CAN ALSO CONTRIBUTE TO FORWARD-LOOKING COMPATIBILITY: AS NEW TYPES OF INTELLIGENCE EMERGE, THEY CAN ALL ACCESS THE SAME BOTTOM NETWORK. BECAUSE OF THE INTEROPERABLE, OPEN-SOURCE, DECENTRIZED AND OFTEN MORE UPGRADED STRUCTURES OF THE BLOCK CHAINS, THEY ARE BETTER SUITED TO THE CHANGES BROUGHT ABOUT BY FUTURE AI INNOVATIONS。

A number of companies are now building a chain of intelligent interactions. In the case of Halliday, for example, the company recently launched an agreement to provide a standardized cross-chain structure for AI workflows and interactions, while incorporating protection mechanisms at the protocol level to ensure that AI does not act beyond the user ' s intent. On the other hand, the Catena, Skyfire, and Nevermind projects use block chains to support automatic settlements between intelligent bodies, making AI-to-AI pay without the intervention of anyone. Similar systems are emerging and Coinbase has begun to provide infrastructure support for such development。

6. AL "CLIMATE ENCODING" APPLICATION TO SYNCHRONIZE

Sam Broner and Scott Duke Kominers: Generating AI revolution made it easier to build software than ever before. Encoding speed increases the order of magnitude and, more importantly, can be done directly through natural languages, allowing inexperienced developers to reset existing programs and even build new applications from scratch。

However, AI Auxiliary Encoding, while creating new opportunities, also brings with it a large number of entropys within and across processes. The so-called vibe coding will abstract the complex dependency behind the software: This is why, when the bottom source reservoir or input changes, the program may be exposed to functional and safety risks. At the same time, when people use AI to create highly personal applications and work streams, the interface between them and other systems becomes more difficult. In fact, even if two vibe-coded programs perform almost identical tasks, they may operate in a completely different logic from the output structure。

TRADITIONALLY, STANDARDIZATION TO ENSURE CONSISTENCY AND COMPATIBILITY HAS BEEN UNDERTAKEN BY DOCUMENT FORMATS, OPERATING SYSTEMS AND LATER SHARED SOFTWARE INTEGRATION WITH API. BUT IN A WORLD WHERE SOFTWARE EVOLVES IN REAL TIME, DEFORMS AND BRANCHES, THE STANDARDIZATION LAYER MUST HAVE BROAD ACCESSIBILITY, CONTINUOUS SCALABILITY AND USER CONFIDENCE. IN ADDITION, AI ALONE DOES NOT SOLVE THE PROBLEM OF INCENTIVE MECHANISMS - THAT IS, HOW DEVELOPERS ARE MOTIVATED TO BUILD AND MAINTAIN LINKS BETWEEN THESE SYSTEMS。

Block chains can address these two challenges simultaneously, and they can provide an agreed synchronous layer that is embedded in a customized software build and can be dynamically updated as the environment changes to ensure compatibility across systems。

In the past, large enterprises may need to pay millions of dollars to systems integrators such as Deloitte to customize an example of Salesforce. Today, an engineer may only need one weekend to build a custom interface for sales data viewing. However, as the number of customized software continues to grow, developers will need help to ensure that these applications remain synchronized and usable。

THIS IS SIMILAR TO TODAY'S OPEN SOURCE LIBRARY DEVELOPMENT MODEL, BUT IT IS DIFFERENT THAT THE SYNCHRONOUS LAYER DOES NOT RELY ON CYCLICAL RELEASES, BUT IS CONSTANTLY UPDATED — AND IS ACCOMPANIED BY INCENTIVES. AND BOTH CAN BE ACHIEVED MORE EASILY THROUGH ENCRYPTION. AS WITH OTHER AGREEMENTS BASED ON BLOCK CHAINS, SHARED SYNCHRONOUS LAYERS CAN INSPIRE SUSTAINED INVESTMENT OF RESOURCES FOR IMPROVEMENT. DEVELOPERS, USERS (AND THEIR AI SMARTS) AND OTHER USERS CAN BE MOTIVATED BY INTRODUCING, USING, OR ITERATIVE NEW FUNCTIONALITY AND INTEGRATION SCHEMES。

on the other hand, shared ownership also allows all users to have an interest in the overall success of the agreement, thus creating disincentives for behavioural deviations. just as microsoft will not easily undermine the .docx file format standards, as it will have a wide negative impact on its users and brands; the co-owners of the synchronized layer will also suffer from their own interests rather than introducing awkward or malicious codes into the agreement。

LIKE ALL PREVIOUS SOFTWARE STANDARDIZATION STRUCTURES, THERE IS ALSO A STRONG NETWORK EFFECT POTENTIAL. AS THE SOFTWARE GENERATED BY AI MOVES TOWARDS A "COLD WAR OUTBREAK", THE DIVERSITY, ISOMERIZATION SYSTEMS THAT NEED TO MAINTAIN COMMUNICATION WITH EACH OTHER WILL GROW EXPONENTIALLY. IN SHORT: THE ATMOSPHERE CODE CANNOT BE SYNCHRONIZED BY THE ATMOSPHERE ITSELF; ENCRYPTION IS THE ANSWER。

7. Supporting the microfinance system for income sharing

Liz Harkavy: AI smarts and tools like ChatGPT, Claude, Copilot provide a more accessible way to access information in the digital world. But for good or bad, they are also destabilizing the economic structure of open Internet. This trend has emerged - for example, as students are increasingly using the AI tool, educational platforms are experiencing a marked drop in flows; at the same time, several media outlets in the United States are also prosecuting OpenAI for copyright infringement. If the incentive system cannot be realigned, we may see a further closure of the Internet, an increase in payment walls and a steady decline in content creators。

POLICY INSTRUMENTS CERTAINLY EXISTED, BUT TECHNICAL SOLUTIONS WERE EMERGING AS JUDICIAL PROCEEDINGS PROGRESSED. ONE OF THE MOST POTENTIALLY (AND AT THE SAME TIME THE MOST TECHNICALLY DIFFICULT) PROGRAMMES IS TO EMBED "INCOME-SHARING" INTO THE BOTTOM OF THE INTERNET. WHEN AN AI-DRIVEN OPERATION ULTIMATELY LEADS TO A SALE, THE CONTENT CREATOR WHO PROVIDES A SOURCE OF INFORMATION FOR THE DECISION SHALL RECEIVE THE SHARE OF PROCEEDS. UNION MARKETING ECOLOGY IS ALREADY DOING SIMILAR ATTRIBUTION TRACKING AND BENEFIT SHARING; A MORE ADVANCED SYSTEM CAN AUTOMATICALLY TRACK AND REWARD ALL CONTRIBUTORS THROUGHOUT THE INFORMATION CHAIN. IT IS CLEAR THAT THE BLOCK CHAIN CAN PLAY A KEY ROLE IN TRACKING THE "SOURCE CHAIN"。

However, to achieve such a system, new infrastructure is needed — in particular: micro-payments systems that can handle very small amounts between multiple sources; attribution agreements that can fairly assess the value of different contributions; and governance models that ensure transparency and fairness。

Many of the existing block chain tools, such as the various rollup, the L2 network, the AI primary financial institution Catena Labs, and the Financial Infrastructure Agreement 0xSplits, have demonstrated their potential to achieve near-zero-cost trades and more fine-scale payments。

The block chain can make it possible, through a variety of mechanisms, for intelligently led advanced payment systems:

Nanopayment: Distinguished between multiple data providers to allow single-user interaction to automatically trigger micropayments for all contributing sources, to be executed by smart contracts。

Smart contract: An enforceable "ex post payment" can be triggered automatically after the completion of the transaction, providing transparent and traceable compensation to those sources of content that influence purchasing decisions。

programmable to pay splits: To enable the distribution of proceeds to be enforced by code rather than relying on centralized agency decisions, thus creating a financial relationship without trust between automated intelligence bodies。

As these emerging technologies mature, they will build a completely new media economy model, from creators to platforms and users to capture the whole value chain。

8. Area chains as intellectual property and traceability registration systems

Scott Duke Kominers: The emergence of the Generable AI makes it urgent to establish an efficient and programmable intellectual property (IP) registration and tracking mechanism - both to ensure the accuracy of traceable sources and to support new business models generated by IP visits, sharing and secondary creations. The existing IP framework, which relies on costly intermediaries and ex post enforcement mechanisms, is clearly incompetent in an age when AI can consume content instantaneously and generate transformations with one key。

WHAT WE NEED IS AN OPEN, PUBLIC REGISTRATION SYSTEM THAT PROVIDES CLEAR PROOF OF OWNERSHIP TO CREATORS AND USES LOW THRESHOLDS AND EFFICIENT -- AND WITH WHICH AI CAN INTERACT DIRECTLY WITH OTHER WEB APPLICATIONS. THE BLOCK CHAIN IS WELL SUITED TO ASSUME THIS ROLE: IT ALLOWS CREATORS TO REGISTER THEIR IP WITHOUT RELYING ON INTERMEDIARIES AND PROVIDES IRREMOVABLE RETROACTIVE PROOF; IT ALSO ALLOWS THIRD-PARTY APPLICATIONS TO EASILY IDENTIFY, AUTHORIZE AND INTERACT WITH THESE IP ASSETS。

OF COURSE, PEOPLE REMAIN CAUTIOUS ABOUT THE OVERALL CONCEPT OF WHETHER TECHNOLOGY REALLY PROTECTS INTELLECTUAL PROPERTY. AFTER ALL, THE FIRST TWO GENERATIONS OF THE INTERNET — INCLUDING EVEN THE CURRENT AI REVOLUTION — ARE OFTEN ASSOCIATED WITH A DECLINE IN IP PROTECTION. ONE REASON FOR THIS IS THAT MANY EXISTING IP BUSINESS MODELS EMPHASIZE "EXCLUSION OF DERIVATIVES" RATHER THAN INCENTIVE, MONETIZATION OF DERIVATIVES. A PROGRAMMABLE IP INFRASTRUCTURE NOT ONLY ALLOWS CREATORS, CONCESSIONAIRES AND BRANDS TO CLEARLY ESTABLISH THEIR IP OWNERSHIP IN DIGITAL SPACE, BUT ALSO FOSTERS A NEW BUSINESS MODEL CENTRED ON SHARING IP FOR GENERATING AI AND DIGITAL APPLICATIONS. IN A SENSE, IT WOULD GENERATE AN AI THREAT TO CREATIVE WORK AND TURN IT INTO A NEW OPPORTUNITY。

In the early stages of the NFT, we have seen creators try to use new models, such as building brand network effects through the CC0 in the Taifeng, to achieve value deposition. Recently, we have seen infrastructure providers begin to build standardized, portfolioable IP registration and licensing agreements, or even introduce specialized block chains (e.g., Story Protocol). Some artists have begun to use Alias, Neuura, Titles and others to authorize their styles and works to support creative mixing. At the same time, the Science-Fiction Series of Incension Emergence engages fans in co-creation and role-setting and records every creative contribution through the chain registration system on Story。

Web reptiles that can compensate content creators

Carra Wu: The most productive-market-compatible AI smarts of the moment are not the types of smarts used for programming or entertainment, but web-based reptiles that can access the Internet, collect data and judge which links should follow。

According to some estimates, nearly half of the current Internet traffic has come from non-human sources. Robots often ignore robots.txt documents — a standard that should have informed the automated reptiles whether they are allowed access or not, but in reality they are almost non-binding — and use captured data to strengthen the core moat of the world's largest technology company. Worse still, the website eventually has to bear the cost of these "uninvited visitors" and pay for bandwidth and CPU resources to deal with the growing number of anonymous reptiles. In response, companies such as Cloudflare and other CDNs provide interdiction services. All of this constitutes a "close" system that should not have existed。

AS WE HAVE POINTED OUT BEFORE, THE ORIGINAL CONTRACT OF THE INTERNET — THE ECONOMIC SYNERGY BETWEEN CONTENT CREATION BY CONTENT CREATORS AND THE DISTRIBUTION OF CONTENT BY THE PLATFORM — IS GRADUALLY BREAKING DOWN. THIS TREND IS ALREADY REFLECTED IN THE DATA: IN THE PAST 12 MONTHS, WEBSITE OPERATORS HAVE STARTED TO BLOCK AL REPTILES ON A LARGE SCALE. IN JULY 2024, ONLY ABOUT 9% OF THE WORLD ' S TOP 10,000 WEBSITES BLOCKED AI REPTILES, WHICH NOW STANDS AT 37%. THIS PROPORTION WILL CONTINUE TO RISE AS MORE WEBSITE OPERATORS MATURE THEIR TECHNOLOGY AND USERS BECOME MORE DISSATISFIED。

So, what happens if we stop paying CDNs to cut them off by "one-size-fits-all" and try a middle path? In other words, AI reptiles no longer "hiking" but pay for the right to access data. Here, the block chain works: in this scenario, each web-based reptile intelligence holds a certain number of encrypted assets and consults on the chain with each website's "doorguard intelligence" or pay-wall protocols through a 402 protocol. (Of course, the challenge lies in the robots.txt, the "robots exclusion criteria" that have been deeply embedded in the operating models of Internet enterprises since the 1990s. To change this, large-scale collaboration or support from CDNs such as Cloudflare is required. I'm not sure

At the same time, human users can prove themselves to be real through World ID (see above) and thus receive free access. As a result, content creators and website operators can be compensated for the fact that data are collected by AI, while human users continue to enjoy the Internet as a free flow of information。

10. Privacy protection-type advertising that is accurate and does not "shock"

Matt Gleason: AI has begun to influence our online shopping, but what if we can actually "use" the ad we see every day? There are many reasons why people do not like advertising: advertising that is not about themselves is pure noise; at the same time, not all "personalization" is good. High-directed advertising, which relies on a large number of consumer data, can make people feel violated; other applications attempt to realize a realization through "compulsory viewing advertising" (e.g., non-jumpable advertising in a serial platform or game level)。

ENCRYPTION TECHNOLOGY CAN HELP TO IMPROVE THESE ISSUES AND PROVIDE AN OPPORTUNITY TO RETHINK THE ADVERTISING SYSTEM. WHEN AI INTELLIGENTS COMBINE WITH BLOCK CHAINS, THEY ARE ABLE TO CUSTOMIZE ADVERTISING ACCORDING TO THE PREFERENCES SET BY THE USER, SO THAT ADVERTISING IS NEITHER IRRELEVANT NOR OVERLY "GENTLE." MORE IMPORTANTLY, IN THE PROCESS, USER DATA WILL NOT BE EXPOSED GLOBALLY, AND USERS WILLING TO SHARE DATA OR INTERACT WITH ADVERTISING WILL BE COMPENSATED。

Achieving this model requires a number of technical foundations:

Low-cost digital payment systemIn order to compensate users for their advertising interactions (watching, clicking, converting), enterprises need to send large amounts of small payments. To scale up, this requires systems with high speed, high throughput and almost zero charges。

Data validation for privacy protectionAI INTELLIGENCE NEEDS TO VERIFY THAT CONSUMERS MEET CERTAIN DEMOGRAPHIC CHARACTERISTICS. THE ZERO KNOWLEDGE CERTIFICATE (ZKP) CAN COMPLETE SUCH CERTIFICATION WITHOUT DISCLOSING SPECIFIC PRIVACY INFORMATION。

New incentive model: if the internet uses a monetization based on micropayments (e.g. & lt; $0.05 per interactive & lt;) the user can voluntarily choose to watch the ad to obtain compensation, thereby transforming the current "data extraction" model into a "user participation model"。

FOR DECADES, PEOPLE HAVE BEEN TRYING TO MAKE ADVERTISING MORE "RELEVANT" — ONLINE AND OFFLINE. RECONSIDERING ADVERTISING FROM THE PERSPECTIVE OF ENCRYPTION TECHNOLOGY AND AI CAN MAKE IT REALLY USEFUL, MANAGEABLE AND OPTIONAL. FOR BUILDERS AND ADVERTISERS, THIS MEANS MORE SUSTAINABLE AND CONSISTENT INCENTIVE STRUCTURES; FOR USERS, IT PROVIDES A RICHER WAY TO DISCOVER INFORMATION AND EXPLORE THE DIGITAL WORLD。

In the end, it would not only add value to advertising posts, but could also destabilize the deep-rooted and “squeeze”-centred advertising economy model today, replacing it with a more people-centred system: In it, the user is no longer a "sold product" but a real participant。

11. AI PARTNERS OWNED AND CONTROLLED BY USERS

Guy Woollet: Today, many people spend more time on equipment than on online communication, which is increasingly being used to interact with the AI model or with content inspired by AI. These models already provide a form of "accompaniment" — whether for entertainment, access to information, satisfaction of the interests of the masses, or as an educational tool for children. It is easy to imagine that, in the near future, AI partners, who are accompanied by education, medical care, legal advice and even daily emotions, will be one of the main ways in which humanity interacts。

FUTURE AI PARTNERS WILL BE ABLE TO BE PATIENT AND IN-DEPTH TAILORED TO INDIVIDUALS AND THEIR USE. NOT ONLY ARE THEY HELPERS OR ROBOT SERVANTS, THEY CAN ALSO BE HIGHLY VALUED BY USERS. THEREFORE, THE QUESTION FOLLOWS: WHO IS OWNERSHIP AND CONTROL OF THESE RELATIONSHIPS — USERS, COMPANIES AND OTHER INTERMEDIARIES? IF YOU HAVE BEEN CONCERNED ABOUT SOCIAL MEDIA CONTENT DEVELOPMENT AND CENSORSHIP OVER THE PAST DECADE, IT WILL BECOME MORE COMPLEX AND PERSONAL IN THE FUTURE。

THE VIEW THAT "THE ANTI-CENSORSHIP HOSTING PLATFORM, SUCH AS BLOCK CHAINS, MAY BE THE BEST WAY TO BUILD AN UNCENSORABLE, USER-CONTROLLED AI" HAS BEEN WELL DISCUSSED. WHILE USERS ARE ABLE TO RUN LOCAL MODELS THEMSELVES AND PURCHASE GPUS, FOR MOST, THIS IS EITHER TOO COSTLY OR TOO HIGH A TECHNICAL THRESHOLD。

Although there is still a distance from full coverage of AI partners, the technology is fast maturing: the text chat type AI has become extremely natural; visualization of virtual images is increasing; and block chain performance continues to improve. If we want to make the "unreviewable AI partners" really easy to use, we need to rely on a better user experience of encryption applications (UX). Fortunately, wallets such as Phantom have made the interactivity of block chains simpler, while technologies such as embedded wallets, Passkey, and abstract accounts have made self-custody easy for users without having to keep their own notes. At the same time, high-intensity, trustless computing systems based on optimism and ZK co-processors will enable us to establish meaningful and sustainable long-term relationships with digital partners。

In the near future, the focus of public discussion will shift from "when there will be real digital partners and virtualization" to "who will control them and how they will be controlled"。

Original Link