Compiled by | Saoirse, Foresight News
The economic logic of the internet has quietly changed. As the open web gradually shrinks into an 'input command bar,' we must consider: will artificial intelligence lead us toward an open internet or into a maze of new payment barriers? Will control lie with large centralized enterprises or with a broad user base?
This is precisely where Crypto comes in. We have discussed the intersection of AI and Crypto multiple times — in short, blockchain is a new paradigm for reconstructing internet service architecture, enabling the creation of decentralized, trusted, neutral, and user-owned web systems. By redefining the economic rules that support existing systems, blockchain provides an effective way to counteract the centralization trends in the AI field, achieving a more open and resilient internet ecology.
The concept of mutually empowering Crypto and AI systems has long existed, but the ways in which they combine have always lacked clear definitions. Some cross-domain areas (such as 'human identity' verification in the context of the proliferation of low-cost AI tools) have attracted the attention of developers and users, while other application scenarios may take years or even decades to materialize. Therefore, this article will share 11 cross-domain application scenarios of AI and Crypto, aiming to promote relevant discussions: exploring the potential possibilities and challenges of the convergence of AI and Crypto, and envisioning more innovative directions. These scenarios are all based on the current technological level, covering diverse fields from massive micro-payment processing to ensuring that humans maintain dominance in future AI interactions.
1. Persistent data and contextual environments in AI interactions
Written by: Scott Duke Kominers
The development of generative AI heavily depends on data, but in many application scenarios, context (i.e., interaction-related state and background information) is as crucial as data, if not more.
Ideally, whether for agents, large language model interfaces, or other applications, AI systems should remember various details about users, such as their type of work, communication style, preferred programming languages, etc. However, in reality, users often need to reset these contexts in different sessions within the same application (for example, starting a new ChatGPT or Claude session), let alone switch between different systems. Currently, the context of a certain generative AI application is almost impossible to transfer to other applications.
With blockchain technology, AI systems can transform key contextual elements into persistent digital assets, allowing them to load at the start of conversations and enabling seamless transfer across different AI platforms. Moreover, due to its characteristics, blockchain may be the only solution that simultaneously meets the requirements of 'backward compatibility' and 'interoperability.'
This application is particularly applicable in AI-mediated gaming and media domains — user preferences (from difficulty settings to key bindings) can remain consistent across different games and scenarios. But the true value lies in knowledge application scenarios (where AI needs to understand users' knowledge reserves and learning patterns) and specialized AI applications (like programming assistance). Of course, some companies have developed custom bots for specific business contexts, but in such scenarios, context often cannot be transferred across systems, even among different AI tools within the same company.
Institutions have only just begun to recognize this problem, and the current universal solution is custom bots with fixed backgrounds. However, context migration among users within the platform has already begun to show signs off-chain: for instance, on the Poe platform, users can rent out custom bots to others.
Once such scenarios are on-chain, our interacting AI systems will be able to share a context layer that includes all critical elements of digital activities. They will immediately understand user preferences and optimize the user experience more precisely. Conversely, like an on-chain intellectual property registration system, allowing AI to reference persistent on-chain contexts also creates possibilities for new market interactions around prompts and information modules — users can directly license or commercialize their own expertise while retaining data control. Of course, shared context will also give rise to many possibilities we have yet to foresee.
2. Universal identity system for agents
Written by: Sam Broner
Identity (i.e., 'the authoritative record of the essential attributes of something') is the underlying architecture supporting today's digital discovery, aggregation, and payment systems. Since platforms have enclosed this architecture within ecological walls, the identity in the eyes of users has become part of product functionality: Amazon assigns unique identifiers (ASIN or FNSKU) to products, centrally displays products, and assists users in discovery and payment; Facebook operates similarly, where user identity is the core foundation of its information flow and the entire application’s discovery functionality, including product listings, native posts, and paid advertisements.
As AI agents evolve, this dynamic is about to change. When more enterprises adopt agents (for customer service, logistics management, payment processing, etc.), their platforms will no longer be limited to single-interface applications but will span multi-platform ecosystems, accumulating deep context and performing more diverse tasks for users. However, if agent identities are bound only to a single market, they will lose availability in other key scenarios (such as email threads, Slack channels, and other products).
Thus, agents require a single, portable 'digital passport.' Without this passport, it would be impossible to determine how to pay agents, verify their version information, inquire about their functional attributes, know their service targets, or trace their reputation across different applications and platforms. Agent identities need to encompass multiple functions such as wallets, API registries, update logs, and social proofs to ensure that any interface (email, Slack, or other agents) can interpret and interact according to a unified standard. Without 'identity' as shared information, every system integration would need to build the underlying architecture from scratch, and discovery mechanisms would always be in a temporary state, causing users to lose contextual information each time they switch channels or platforms.
We have the opportunity to design agent infrastructure from foundational principles. So, how do we build a trustworthy neutral identity layer that is more refined than a DNS record? Agents should not repeat the mistakes of 'binding identity and discovery, aggregation, and payment functions' in monolithic platforms, but should be able to accept payments and display functionalities across multiple ecosystems without being locked into a specific platform. This is precisely where the value of the intersection of Crypto and AI lies — the permissionless composability provided by blockchain networks can help developers build more practical agents and better user experiences.
Overall, vertically integrated solutions (like Facebook or Amazon) currently offer a better user experience. One of the inherent challenges in creating excellent products lies in ensuring that various components work collaboratively from top to bottom. However, the cost of this convenience is high, especially as the costs of building agent aggregation, marketing, commercialization, and distribution software continue to decline, and the coverage of agent applications continues to expand. Although there is still work to be done to reach the user experience level of vertically integrated platforms, building a trustworthy and neutral identity layer for agents will enable entrepreneurs to take control of their 'digital passports' and promote innovation and exploration in distribution and design.
3. Forward-compatible 'human identity' proof mechanism
Written by: Jay Drain Jr., Scott Duke Kominers
As AI technologies permeate various online interaction scenarios (including deep fakes and social media manipulation), determining 'whether one is interacting with a real human online' is becoming increasingly difficult. The collapse of this trust system has already occurred — from comment armies on the X platform (formerly Twitter) to bot accounts in dating applications, the boundary between virtual and reality is gradually blurring. In this context, 'human identity' proof has become a core infrastructure of the digital ecosystem.
One way to prove 'I am human' is by utilizing digital IDs (including centralized IDs used by TSA). Digital IDs encompass all elements used for identity verification, including usernames, PINs, passwords, and third-party certifications (such as citizenship or credit ratings). Decentralization demonstrates significant value here: when data is stored in centralized systems, issuers may revoke access, charge additional fees, or implement monitoring; in contrast, decentralized models reverse this logic — users (rather than platform administrators) maintain control of their own identity, making it safer and censorship-resistant.
Unlike traditional identity systems, decentralized 'human identity' proof mechanisms (such as World’s Proof of Human) allow users to autonomously manage their identity information and verify 'human attributes' in a privacy-preserving and trusted neutral way. Just as a driver's license can be used in any region (regardless of when or where it was issued), decentralized 'human identity' proof can serve as a universal underlying protocol across platforms, even applicable to emerging platforms that have yet to be born. In other words, blockchain-based 'human identity' proof possesses forward-compatible characteristics due to the following advantages:
· Portability: Relevant protocols belong to open standards, and any platform can integrate. Decentralized 'human identity' proof can be managed through public infrastructure, controlled autonomously by users, providing complete portability, and compatibility can be achieved on any platform now or in the future;
· Permissionless accessibility: Platforms can autonomously choose to recognize 'human identity' IDs without going through APIs that may discriminate against different use cases.
The challenge in this area is the application of grounded use cases: Although there have not yet been practical-scale 'human identity' proof application scenarios, we expect the adoption rate to accelerate when the number of users reaches a critical mass, early partnerships are formed, and killer applications emerge. Each application adopting a specific digital ID standard will enhance the value of that ID to users, thereby attracting more users to obtain the ID, creating a positive feedback loop (and due to the inherent interoperability of on-chain IDs by design, network effects can accumulate rapidly).
We have seen mainstream consumer applications in gaming, dating, and social media announce partnerships with World ID to help users confirm that they are interacting with real humans (rather than programs) in games, chats, and transactions; this year has also seen the emergence of new identity protocols like Solana Attestation Service (SAS) — although SAS is not an issuer of 'human identity' proof, it allows users to privately associate off-chain data (such as compliant KYC or investment qualifications) with their Solana wallets, helping to build a decentralized identity system. All these signs indicate that the turning point for decentralized 'human identity' proof may not be far off.
The significance of 'human identity' proof lies not only in banning bots but in clearly delineating boundaries between AI agents and human networks. It allows users and applications to distinguish between human and machine interactions, thus creating a higher quality, safer, and more authentic digital experience.
4. Decentralized infrastructure (DePIN) in the AI field
Written by: Guy Wuollet
AI, while a digital service, is increasingly constrained by physical infrastructure. Decentralized infrastructure networks (DePIN) offer a new model for building and operating real-world systems, contributing to the democratization of the computing infrastructure behind AI innovation, making it more economical, resilient, and censorship-resistant.
How can this goal be achieved? The two core challenges facing AI development are computing power supply and chip acquisition. Decentralized computing networks can provide more computing power, while developers are also leveraging DePIN to aggregate idle chip resources from gaming PCs, data centers, and other sources. These computing devices can form a permissionless computing market, creating a fair competitive environment for developing new AI products.
Other application scenarios include distributed training and fine-tuning of large language models, as well as distributed networks for model inference. Decentralized training and inference (utilizing idle computing resources) can significantly reduce costs while providing censorship resistance, ensuring that developers will not have their services terminated by oversized cloud service providers (like centralized cloud service giants).
The problem of a few companies monopolizing AI models has long existed, while decentralized networks help build a more economical, censorship-resistant, and scalable AI ecosystem.
5. Infrastructure and rule framework for interactions between AI agents, terminal service providers, and users
Written by: Scott Duke Kominers
As AI tools continue to improve in solving complex tasks and executing multi-level interaction chains, AI systems will increasingly need to interact with other AI systems without human intervention.
For example, a particular AI agent may need to request specific data related to computing tasks or recruit specialized AI agents to complete specific tasks (like assigning statistical bots to develop and run model simulations or invoking image generation bots when creating marketing materials). AI agents will also create significant value in completing the entire transaction process or other activities on behalf of users, such as searching for and booking flights based on user preferences or discovering and ordering new books of their favorite genre.
Currently, there is no mature universal market for inter-agent interactions; such cross-system queries can mostly only be implemented through explicit API connections or within AI ecosystems that treat agent calls as internal functions.
Overall, most AI agents today operate in siloed ecosystems, where API interfaces are relatively closed and lack architectural standardization. However, blockchain technology can help protocols establish open standards, which is crucial for short-term application implementation; in the long run, this also supports forward compatibility — as new AI agents evolve and emerge, they can access the same underlying network. Due to its interoperable, open-source, decentralized, and often more easily upgradeable architectural characteristics, blockchain can flexibly adapt to the innovative demands in the AI field.
As the market develops, several companies have begun building blockchain infrastructures for interactions between agents: for example, Halliday recently launched related protocols to provide standardized cross-chain architectures for AI workflows and interactions, and offers protective mechanisms at the protocol layer to ensure that AI behavior does not exceed user intent; Catena, Skyfire, and Nevermind use blockchain technology to support automatic payments between AI agents without human intervention. More such systems are in development, and Coinbase has even begun to provide infrastructure support for these explorations.
6. Ensure the synchronization of AI/custom programming applications
Written by: Sam Broner, Scott Duke Kominers
The innovation of generative AI has led to a qualitative leap in software development efficiency: coding speed has increased by several orders of magnitude, and most importantly, it can now be accomplished through natural language — even inexperienced programmers can replicate existing programs or build new applications from scratch.
However, AI-assisted coding creates new opportunities while also introducing significant uncertainty both internally and externally. 'Custom programming' (Vibe coding) abstracts the complex dependency networks underlying software, but this also makes programs vulnerable to functional and security vulnerabilities when changes occur in source libraries and other inputs. Furthermore, as people use AI to create personalized applications and workflows, interactions with others' systems become more difficult — in fact, even if two 'custom programming' programs have the same functionality, their operational logic and output structures may exhibit significant differences.
Historically, ensuring software consistency and compatibility was primarily undertaken by file formats and operating systems, while in recent years, it has relied on shared software and API integration. However, in the new era of real-time software evolution, iteration, and branching, the standardization layer needs to have broad accessibility and continuous upgradability while maintaining user trust. Furthermore, AI alone cannot solve the problem of 'incentivizing people to build and maintain these connections.'
Blockchain technology addresses both of these issues: the protocolized synchronization layer can be embedded in users' customized software architectures and dynamically updated to ensure cross-system compatibility as environments change. Historically, large enterprises might pay millions of dollars to system integrators like Deloitte to customize Salesforce instances. Today, engineers can create customized interfaces to view sales information over the weekend, but as the number of customized applications grows, developers need professional support to maintain the synchronization of these applications. (Note: Salesforce is a customer relationship management (CRM) software service provider established in March 1999 in the United States.)
This is similar to the development model of today's open-source software libraries, but with continuous updates (rather than periodic releases) and incentive mechanisms — both made easier by Crypto technology. Like other blockchain-based protocols, the shared ownership mechanism of the synchronization layer incentivizes all parties to actively invest in improvements: developers, users (and their AI agents), and other consumers can be rewarded for introducing, using, and optimizing new features and integrations.
Conversely, shared ownership tightly binds all users to the overall success of the protocol, forming a buffer mechanism against malicious behavior — just as Microsoft would not easily destroy the .docx file standard (as it would have a chain reaction effect on users and brands), the co-owners of the synchronization layer are also unlikely to introduce inefficient or malicious code into the protocol.
As with all software standardized architectures we have seen, there exists significant potential for network effects here. As the 'Cambrian explosion' of AI coding software continues to evolve, the heterogeneous systems network that needs to maintain communication will expand exponentially. In short, 'custom programming' requires not only 'coding styles' but also Crypto technology to maintain system synchronization.
7. Micro-payment systems supporting revenue sharing
Written by: Liz Harkavy
AI agents and tools like ChatGPT, Claude, and Copilot offer new ways to navigate the digital world conveniently, but whether good or bad, they are shaking the economic foundations of the open internet. We have seen tangible manifestations of this trend — for instance, educational platforms have seen a sharp decline in traffic due to significant AI tool usage by students, and several U.S. newspapers are suing OpenAI for copyright infringement. Without readjusting the incentive mechanisms, we may face an increasingly closed internet: rising paywalls and a decrease in content creators.
Of course, policy solutions always exist, but as they progress through judicial processes, a series of technological solutions are also emerging. The most promising (and technically challenging) solution may be to embed revenue-sharing mechanisms into network architectures: when AI-driven behavior facilitates a transaction, the content sources that provide information supporting that decision should receive respective shares. Affiliate marketing ecosystems have been doing similar attribution tracking and revenue-sharing, while more advanced versions could automatically track and reward all contributors in the information chain — blockchain technology can clearly play a key role in tracing this provenance chain.
However, such systems also require new types of infrastructure with additional functionalities — particularly micro-payment systems capable of processing micro-transactions across multiple sources, attribution protocols that fairly assess the value of different contributions, and governance models that ensure transparency and fairness. Many existing blockchain tools (such as Rollups and Layer2, AI-native financial institutions like Catena Labs, and financial infrastructure protocol 0xSplits) have already demonstrated application potential, supporting near-zero-cost transactions and more granular payment splits.
Blockchain can implement complex agent payment systems through various mechanisms:
· Nano payments can be split among multiple data providers, allowing a single user interaction to trigger micro-payments to all contributing sources via automated smart contracts.
· Smart contracts support traceable payments that can be triggered after a transaction is completed, compensating sources of information that informed purchasing decisions in a fully transparent and traceable manner.
· Additionally, blockchain supports complex and programmable payment split distributions, ensuring that revenues are fairly distributed through code-enforced rules rather than relying on centralized decisions, creating trustless financial relationships among autonomous agents.
As these emerging technologies mature, they will create new economic models for the media industry, capturing the entire value creation chain from creators to platforms to users.
8. Blockchain as an intellectual property and provenance registration system
Written by: Scott Duke Kominers
The development of generative AI has created an urgent need for efficient and programmable intellectual property registration and tracking mechanisms — both to clarify rights ownership and to support business models around access, sharing, and re-creation of intellectual property. Existing intellectual property protection frameworks rely on expensive intermediaries and post-factum enforcement measures, unable to adapt to the demands of an era where AI instantaneously consumes content and generates new variants with one click.
What we need is an open public registration system that provides clear proof of ownership, enabling intellectual property creators to efficiently participate in interactions, and allowing AI and other web applications to connect directly. Blockchain technology is the ideal choice: it allows for intellectual property registration without intermediaries, provides tamper-proof provenance proof, and enables third-party applications to easily recognize, authorize, and utilize that intellectual property.
Some are skeptical of the view that 'technology can protect intellectual property' — after all, the first two development eras of the internet (and the ongoing AI revolution) are often associated with the weakening of intellectual property protection. Part of the reason is that many contemporary business models based on intellectual property focus on 'prohibiting derivative works' rather than incentivizing and commercializing derivative creations. However, programmable intellectual property infrastructure not only allows creators, brands, and IP owners to clearly establish ownership in the digital space but also opens the door for 'business models around IP sharing (for generative AI and other digital applications)' — this effectively transforms the main threat of generative AI to creation into an opportunity.
We have seen creators experimenting with new models in the early NFT space: companies leverage NFT assets on Ethereum to support network effects and value accumulation under CC0 branding; recently, infrastructure providers are also building protocols for standardized and composable IP registration and authorization (like Story Protocol) or even dedicated blockchains. Some artists have begun using these tools to license their styles and works for creative re-creation through agreements like Alias, Neura, and Titles. Invention’s Emergence series invites fans to co-create a sci-fi universe and its characters, with the blockchain registry built on Story Protocol tracking the creator attribution of each element.
9. Mechanisms for compensating content creators from web crawlers
Written by: Carra Wu
Today, the market's most in-demand AI agents are not programming or entertainment tools but web crawlers — they autonomously browse the web, collect data, and determine scraping sources.
It is estimated that nearly half of current web traffic comes from non-human entities. Crawlers often ignore the robots.txt protocol (which is supposed to inform automated crawlers whether they are allowed to access a website but has weak actual enforcement), and utilize the data they scrape to strengthen the market barriers of tech giants. Worse, websites have to foot the bill for these uninvited guests, bearing the costs of providing bandwidth and CPU resources for massive unidentified crawlers. In response, CDNs (content delivery networks) like Cloudflare offer blocking services, but this is merely a patchwork solution that should not exist.
We have pointed out that the native protocols of the internet (economic agreements between content creators and distribution platforms) may collapse, and data is validating this trend. In the past 12 months, website owners have massively blocked AI crawlers: in July 2024, only 9% of the top 10,000 websites globally prohibited AI crawlers, while that number has now risen to 37%, and as more website operators upgrade their technologies and user dissatisfaction increases, this number is expected to climb further.
Is it possible to find a compromise solution without relying on CDNs to completely block suspected bot access? AI crawlers should not exploit systems designed for human traffic for free but should pay for data scraping access. This is precisely where blockchain comes in: each web crawler agent can hold Crypto and negotiate on-chain with each website's 'access agents' or paywall protocols through the x402 protocol (of course, the challenge lies in the fact that the robots.txt protocol has been deeply embedded in Internet business logic since the 1990s, requiring large-scale collaboration or the involvement of CDNs like Cloudflare to break through).
Meanwhile, humans can prove their identity via World ID (see Chapter 3) to access content for free. In this model, content creators and website owners can receive compensation during AI dataset collection, while human users can still enjoy an 'information-free' internet.
10. Privacy-preserving personalized advertising
Written by: Matt Gleason
AI has begun to influence online shopping experiences, but what if the ads people see daily could be 'truly useful'? The reasons people dislike ads are obvious: ineffective ads are just noise, while overly precise AI ads based on vast consumption data seem intrusive to privacy. Other applications profit by imposing 'non-skippable ads' on content limitations (like streaming services or game levels).
Crypto offers the possibility of reconstructing advertising models. Personalized AI agents combined with blockchain can find a balance between 'irrelevant ads' and 'overly precise ads,' delivering ads based on user-defined preferences. More importantly, this model does not require exposing global user data and can directly compensate users who actively share data or interact with advertisements.
To achieve this goal, the following technical requirements must be met:
· Low-fee digital payments: To compensate users for their ad interactions (views, clicks, conversions), companies need to frequently send small payments, which requires the system to have high-speed processing capabilities and almost zero transaction fees;
· Privacy-preserving data verification: AI agents need to prove that users meet specific demographic attributes, and zero-knowledge proofs can complete attribute verification while protecting privacy;
· Incentive model: If the internet adopts a micro-payment-based profit model (e.g., individual interactions costing less than $0.05, see Chapter 7), users could autonomously choose to 'watch ads for small rewards,' transforming the existing 'exploitative' model into a 'participatory' model.
For decades, online (and offline for hundreds of years) advertising has pursued 'relevance.' Reconstructing advertising from the perspectives of Crypto and AI will ultimately make it more practical — customized yet unobtrusive, benefiting all parties: for developers and advertisers, unlocking more sustainable and incentive-compatible business models; for users, gaining more pathways to explore the digital world.
This will not only enhance the value of advertising slots but may also disrupt today's deeply entrenched 'exploitative' advertising economy, building a more human-centered system — users are no longer commodities to be traded but are active participants.
11. AI companions owned and controlled by humans
Written by: Guy Wuollet
Today, people spend more time on devices than in offline interactions, increasingly used for interacting with AI models and AI-generated content. These models have begun to provide companionship value, whether for entertainment, information retrieval, satisfying niche interests, or educating children. It is not hard to imagine that in the near future, AI companions for education, healthcare, legal consulting, and emotional companionship will become mainstream interaction modes.
Future AI companions will possess infinite patience and be tailored to individual needs — they will no longer be merely tools or robotic servants but may become highly valued relationships. Therefore, the question of 'who owns and controls these relationships' is crucial (is it the user, or intermediaries like companies?). If you have been concerned about content filtering and censorship on social media over the past decade, this issue will become even more complex and personal in the future.
The notion that 'censorship-resistant blockchain hosting platforms are the most viable path to achieving user control over AI' has been discussed multiple times (as mentioned earlier). In theory, individuals could run device-end models or purchase GPUs themselves, but most people either cannot afford it or lack the technical capability.
Although the popularization of AI companions will take time, related technologies are rapidly iterating: anthropomorphic text interaction companions are already quite mature, visual avatar technology has significantly advanced, and blockchain performance continues to improve. To ensure that censorship-resistant companions are user-friendly, we need to rely on better user experiences to realize crypto-driven applications. Fortunately, wallets like Phantom have greatly simplified blockchain interactions, and embedded wallets, cryptographic keys, and account abstraction technologies allow users to hold self-custody wallets without remembering seed phrases. High-throughput trusted computing technologies based on Optimistic and ZK co-processors will also help build deep and lasting relationships with digital companions.
In the near future, the focus of discussion will shift from 'when will realistic digital companions appear' to 'who can control them.'
