Abstract: Cycles and narratives have always been the core topics of the global crypto market. In the past, the industry used Bitcoin halving as a reference to perceive cycles and explore major narrative trends. However, after the approval of Bitcoin and Ethereum spot ETFs, the crypto market is highly coupled with the global financial market trends, and the variables affecting the crypto market trends are increasing.
In the context of soaring chaos values, it is very important to perceive the periodicity more clearly and discover the future narrative trends. As innovative narrative catchers, investment institutions have always been relatively cutting-edge. In view of this, OKX specially planned the Crypto Evolution column, inviting mainstream crypto investment institutions around the world to systematically output topics such as the periodicity of the current market, the direction of the new round of narratives, and the subdivision of popular tracks, in order to stimulate discussion.
The following is the third issue, which is jointly discussed by OKX Ventures, Polychain and Delphi Digital around topics such as The Integration of AI and Crypto. I hope their insights and opinions will inspire you.
About OKX Ventures
OKX Ventures is the investment arm of OKX, a leading crypto asset trading platform and Web3 technology company, with an initial capital commitment of $100 million. It focuses on exploring the best blockchain projects globally, supporting cutting-edge blockchain technology innovation, promoting the healthy development of the global blockchain industry, and investing in long-term structural value. Through its commitment to entrepreneurs who support the development of the blockchain industry, OKX Ventures helps build innovative companies and brings global resources and historical experience to blockchain projects.
About Polychain Capital
Polychain Capital is an investment company dedicated to investing in blockchain technology and decentralized finance. Since its establishment in 2016, it has focused on supporting transformative projects that use blockchain to disrupt traditional industries and create innovative financial systems. With a deep understanding of the crypto space, Polychain Capital supports visionary entrepreneurs and cutting-edge technologies to drive the growth and adoption of decentralized networks and protocols.
About Delphi Digital
Delphi Digital is a research-driven company dedicated to advancing the understanding and development of the growing digital asset market. Delphi Digital supports the ecosystem through four business lines: Delphi Research, Delphi Ventures, Delphi Creative and Delphi Labs.
1. When Crypto meets AI
OKX Ventures: The development of AI technology is currently highly dependent on the promotion of giants, such as OpenAI, Google, Nvidia, etc. Nvidia controls the electricity of the entire AI era, while OpenAI and Google control the most core data and technical solutions. This centralized and highly dependent state of giants will limit the innovation and development of the industry. The decentralized and permissionless characteristics of Crypto can break the shackles of giants, promote technological innovation to a certain extent, and bring new prosperity to the industry.
Currently, the more common scenarios include computing power, data, models and applications.
• Computing power
Distributed/decentralized computing power markets such as io.net and Prodia can break the computing power monopoly of giants by utilizing idle computing power in the global market. We are very much looking forward to seeing what kind of chemical reaction will occur when the total supply of distributed computing power exceeds the centralized computing power supply at a certain point in the future. In addition, due to the scarcity and high returns of AI computing power assets, computing power RWA projects such as Compute Labs have also emerged, creating an AI-Fi ecosystem by tokenizing computing power assets and developing related derivatives.
• data
Cryptos economic model can well motivate users to participate in the field of AI data. For example, various Depin projects can motivate users to participate in data contribution, data labeling or data verification through token economic models, providing data sources for AI model training; Space and Time combines tamper-proof on-chain and off-chain data through proof of SQL to create a verifiable computing layer for the combination of AI and blockchain; 0g.ai has built a scalable data availability layer and storage system; in addition, Cryptos privacy protection features can better realize the security and privacy of user data. For example, Flock.io, Privasea.ai, etc. all emphasize the importance of user data privacy protection during model training.
• Model
The open model market is expected to break the model monopoly of technology giants. Users can not only support the training and reasoning of AI models by providing computing resources, but also provide data or models and interact directly through network protocols. In addition, distributed model training is currently a difficult point. We especially hope to see technological breakthroughs in distributed model training, and hope to see entrepreneurial teams fill this blank in the near future.
• Application
At the application level, the combination of AI and Crypto can better enable creators to generate content. Users can independently build virtual characters and chatbots with customized personalities, such as Myshell. Users can build their own AI Smart Agent by uploading data training models. At the same time, data providers and model trainers can also benefit from the development of the platform, thereby forming a positive data flywheel.
Polychain: The AI space is shifting from closed-source models to sophisticated open-source solutions. While this change has democratized access to AI capabilities, it has also created new challenges, especially in terms of value capture for model creators. The financialization of open-source models is an innovative area at the intersection of crypto and AI, where blockchain technology combines open source, ownership, and verifiability to lay the foundation for value accumulation. Oras Initial Model Offering (IMO) demonstrated how tokens can represent AI models, allowing token holders to be rewarded when the model is profitable. This not only incentivizes open-source development, but also ensures fair compensation for creators and contributors.
In addition to financialization, the convergence of crypto and AI is also driving innovations in public governance and system transparency. As concerns about AI model bias and centralized control increase, blockchain-based solutions provide decentralized training, reasoning, and governance mechanisms, ensuring transparent decision-making processes and community participation.
However, the core of innovation lies in the development of infrastructure. Advances in distributed computing networks, novel data ownership mechanisms, and new token standards make model ownership and revenue sharing possible. These infrastructure developments lay the foundation for complex applications of crypto and AI.
One promising direction is the emergence of AI agents and executable task systems. They can act as an extension of individuals and automatically perform complex tasks, from personalized assistants to advanced automation in decentralized finance. However, their realization depends on the seamless integration of data privacy protection, verifiable computing systems, and infrastructure.
Projects that merge cryptography and AI are in their early stages but are making progress through rapid iterative experimentation. Although best practices have yet to be established, the potential of cryptography to solve AI challenges is becoming increasingly clear.
In the future, we will see more sophisticated applications that combine the advantages of encryption and AI. This fusion will not only lead to a more transparent and accountable system, but will also significantly improve the usability and functionality of AI and blockchain technology. As exploration continues, there will be exciting developments in this area.
2. Crypto+AI Investment Methodology
OKX Ventures: We can answer this question by starting from the current development trend of the track.
• This track is currently moving from hype to reality.
In the past year, a large number of Crypto AI projects have emerged in the market, most of which are concentrated in the infrastructure field, but less in the application level. They are mainly shell projects, lack innovation, have low technical content, and are largely hype. Hype and bubbles are the companion phenomena of the early stage of technological innovation. After the optimization of market resources, we will see real entrepreneurial teams with technical backgrounds entering the Crypto AI field in the future. The market will begin to favor projects that can provide real value, scalability and usability, rather than those that rely solely on hype and marketing.
• From speculation to demand
The market will shift from being driven by speculation to being driven by demand, and the focus will shift from speculation on potential value to actual use and adoption. It is not enough for entrepreneurs to attract investors with narratives alone. The market is now cautious and conservative about pure narrative projects. In the future, projects with real market demand and business income will become a necessary condition for investors to pay, which is also a basic logic for us to layout the Crypto AI track.
Based on the above development trends, we have abstracted the corresponding investment logic, which has three core points:
• Market demand oriented
Many AI startup teams only discovered that the market was not buying their products and that users had no demand for their products after they launched them. The root cause of this is that they did not conduct enough market demand research at the beginning of their business and were not market-oriented, or they mistakenly identified unverified pseudo-demands.
Therefore, when planning the Crypto AI track, we pay special attention to meeting market demand. First, we need to determine in general which track the project belongs to in the Crypto AI field, how big the market capacity will be, how much room the project will have for future development, and what the competitors and competition situation are. Secondly, what problems the project solves and what needs it meets. Even if it is a very small point, solving the market pain point is a feasible idea and direction.
• Not purely narrative
The most criticized thing about the Crypto AI field is that it is narrative-based and has no practical application. It is difficult for us to completely agree with this point of view, but the market is unlikely to pay for pure narratives anymore. Therefore, real business scenarios and business models are particularly important.
The entrepreneurial team must have business income that can support its own survival. Many entrepreneurial teams use NFT/Token sales as their only source of income, which is absolutely not feasible. The entrepreneurial team must have a clear business model and think clearly about where their business income comes from, rather than just relying on narratives to get the market to pay.
• The team needs to have AI background
The popularity of AI quickly ignited the enthusiasm of the Web2 market and VCs. This trend naturally blew into the crypto world. Many crypto startup teams began to take advantage of AI hot spots and package projects, resulting in a large number of Crypto AI projects on the market. However, since the teams had no AI technical background, most of the projects were shell products and had no market competitiveness, and were quickly eliminated by the market. AI has a high technical threshold, especially when Crypto and AI are combined, a deep understanding of both fields is required to effectively combine the two, otherwise it is difficult to be recognized by the market.
In general, the basic idea of investment is to discover market needs and problems in a track with huge market development potential, and find the most suitable team to provide assistance to entrepreneurs, so as to meet market needs and solve problems from 0 to 1.
Polychain: The current landscape of crypto and AI projects is dominated by narrative-driven, which is typical of early-stage transformative technologies. This narrative is not only part of marketing, but also a necessary part of ecosystem development, helping to attract attention, drive community participation and early adoption. However, we recognize that when evaluating these projects, we need to look beyond the narrative and focus on their technical foundations and practical applications. Therefore, our investment strategy is based on in-depth research on crypto and AI technologies and their potential synergies, prioritizing projects with attractive visions, clear market adoption paths and solid technical foundations. Filtering out this noise requires in-depth research.
Currently, the combination of encryption and AI is mainly at the infrastructure level, in areas such as GPU networks, reasoning and intelligent networks, verifiable and private computing, and data management solutions, which will lay the foundation for the next wave of innovation.
In the future, data privacy technologies such as homomorphic encryption, multi-party computing, and zero-knowledge proofs will become important means of protecting AI privacy. Decentralized data markets, verifiable reasoning networks, and AI proxy infrastructure will continue to grow, promoting the popularization of AI functions and building fair, transparent, and efficient systems. The integration of AI and blockchain may bring a new wave of crypto applications, such as AI analysis of decentralized finance, predictive models for asset management, and governance mechanisms for DAOs. Small and efficient models trained on high-quality data sets will also continue to develop, promoting more personalized AI experiences and reducing application friction.
Delphi Digital: Software is eating the world, and AI is eating software. The essence of AI is data and compute, so whoever can most effectively take these two key inputs (infrastructure), coordinate them (middleware), or leverage them to meet user needs (applications) will capture tremendous value.
At present, Delphi Ventures core investment logic focuses on the DeAI ecosystem, and is actively deploying projects at every layer of the DeAI stack.
First, at the infrastructure level, DeAI relies on data and computing, especially through cryptocurrency incentives to efficiently obtain these resources. This is the most challenging but also the most promising part of the stack. Currently, distributed training protocols and the GPU market provide low-cost solutions for large technology companies by coordinating heterogeneous hardware, and the DePIN network occupies an important position in the future intelligent economy with its ability to build hardware networks at low cost.
Second, at the middleware level, DeAI aims to enable efficient composable computing, similar to the Lego model of DeFi. We are particularly optimistic about efficient routing mechanisms (i.e. how to choose the most cost-effective and performant model for the right use case), graph neural networks, coprocessors for scaling data and computing in restricted on-chain environments, and cryptographic-based mechanisms to solve the incentive problem for open source developers. If executed well, DeAI middleware will depict a compelling vision of a modular approach to AI that may eventually surpass the integrated closed-source versions of todays tech giants.
Finally, at the application level, on-chain proxy protocols may be the key to improving the user experience in the crypto space. By connecting computing networks and users, these protocols can not only reduce costs, but also unlock the potential of web3 infrastructure and promote the development of new economic models.
Overall, AI will profoundly change our economic shape. While the current narrative of DeAI may be overly optimistic, the scale of the opportunity is indeed huge. For those with patience and insight, DeAIs true vision of composable computing may prove the value of blockchain itself.
3. Talk about future opportunities
OKX Ventures: Technological breakthroughs and innovations are eternal opportunities.
There is a serious technological monopoly in the field of AI. Data and core technologies are more in the hands of technology giants, and the living space of start-ups is squeezed very small. How to deal with the technological monopoly and cannibalization of technology giants is the primary problem that entrepreneurs need to face. We are very much looking forward to seeing more entrepreneurial teams in the future jump out of the identity of followers, break the monopoly of centralized technology giants through the combination of Crypto AI, achieve technological breakthroughs and innovations, and truly implement narratives and products to meet market demand.
How to avoid leaving the table is a question that entrepreneurial teams need to think about.
• Entrepreneurial teams need to explore the business model and sustainability of their own business. Purely narrative projects are no longer accepted by the market. Entrepreneurial teams need to have stable business income or a clear, feasible and monetizable business model in the future.
• The entrepreneurial team needs to have reasonable financial management and cost control skills to ensure the long-term stable operation of the project. Financial issues are the most common problem for entrepreneurial teams, and many entrepreneurial teams fail due to poor financial management.
• Entrepreneurial teams need to be flexible and maneuverable. Market development is changing rapidly, and a technological breakthrough may cause a group of entrepreneurial teams to disappear. Entrepreneurial teams need to be flexible and adaptable, adjust strategies and directions in a timely manner according to the market, and learn to choose the right time and take advantage of the situation.
Polychain: Currently, the sentiment in the AI and crypto industries is undergoing a significant shift. The attitude of institutions and regulators towards the crypto market has improved, and the approval of Bitcoin and Ethereum ETFs in the United States reflects this change. The increase in mainstream acceptance paves the way for further innovation. At the same time, the field of AI is also undergoing a transformation. Some of the founding members of OpenAI left to promote the realization of the concept of super alignment, which brought new innovation opportunities to AI development and governance, which is in line with the decentralized spirit of crypto projects and forms a unique synergy.
Although the AI field is still in its early stages, demand remains high, but the dominant strategy has not yet been clarified. Combined with the concept of fairer and more open AI, projects that can effectively combine AI and blockchain technology have broad development prospects. The concept of super alignment alleviates concerns about the impact of AI on employment and information authenticity, and also promotes interest in user-owned AI systems. Crypto projects that can promote user ownership and interest alignment have received widespread attention.
However, opportunities come with challenges. The global economy is facing pressures such as conflict, recession, high inflation and high interest rates, which have led to cautious consumption and may affect investment in crypto assets. But this environment may also prompt people to see cryptocurrencies as an alternative to the traditional financial system, and Bitcoin is therefore seen as digital gold and a means of storing value in uncertain times.
Regulatory uncertainty remains an issue. The legal environment for cryptocurrencies and AI varies from region to region, and projects must remain flexible in this uncertainty. Talent scarcity is also a major challenge. There is fierce competition for professional talent in the fields of AI and blockchain, which may affect the progress of project development.
Looking ahead, the current market cycle may become a filter for crypto and AI projects. Projects that can effectively respond to real needs, adapt to regulation and integrate technology will lead the next stage of development in the industry. As the market matures, crypto and AI technologies will develop in a more sustainable and practical direction. In the future, more attention may be paid to user ownership and data rights of AI systems, the development of decentralized AI infrastructure, and the deep integration of AI capabilities with the blockchain ecosystem to promote the emergence of new economic models.
Delphi Digital: The biggest challenge for DeAI is at the infrastructure level, especially the capital intensity required to build basic models and the returns on scale of data and computing.
Big tech companies have a clear advantage here: having built a massive pool of capital using monopoly profits during the second-generation internet era and reinvested it in cloud infrastructure during a decade of low interest rates, they are now attempting to corner the data and computing markets and, in turn, the intelligence market – a key element of AI.
Due to the capital requirements and high bandwidth requirements of large training, super clusters are still the best choice, which provides large technology companies with the most powerful closed-source models and plans to rent them out at monopoly profits and reinvest the proceeds in next-generation technologies. However, AIs moat is shallower than the network effect of web2. The value of cutting-edge models is depreciating rapidly, especially as Meta turns to a scorched earth strategy and invests tens of billions of dollars to develop open source models with SOTA performance such as Llama 3.1.
The commoditization of cutting-edge models is becoming increasingly evident with the rise of low-latency decentralized training methods. This shift shifts competition away from hardware superclusters controlled by large tech companies and toward an environment more conducive to open source and cryptographic software innovation. At the same time, the price of smart technology is falling rapidly.
We have explored the tension between big tech and DeAI in depth in our DeAI series of reports , which interested readers are welcome to read for free.
Given the computational efficiency of the “hybrid of experts” architecture and LLM synthesis/routing, the future may not be a world dominated by 3-5 super models, but an intelligent network of millions of models of different shapes, sizes, and use cases. This brings huge coordination challenges, and blockchain and crypto incentive mechanisms have the potential to solve these problems.
OKX Ventures Disclaimer, please read OKX Disclaimer .
Delphi Digital Disclaimer. Please read the Delphi Digital Disclaimer.
Risk Warning and Disclaimer
This article is for reference only. This article only represents the authors views and does not represent the position of OKX. This article is not intended to provide (i) investment advice or investment recommendations; (ii) an offer or solicitation to buy, sell or hold digital assets; (iii) financial, accounting, legal or tax advice. We do not guarantee the accuracy, completeness or usefulness of such information. Holding digital assets (including stablecoins and NFTs) involves high risks and may fluctuate significantly. You should carefully consider whether trading or holding digital assets is suitable for you based on your financial situation. Please consult your legal/tax/investment professionals for your specific situation. Please be responsible for understanding and complying with local applicable laws and regulations.
