99% of crypto AI entities will fail, only those that are truly useful will survive.
Author: Teng Yan
Translation: Deep Tide TechFlow
On a crisp January morning in 2026, you find a slightly outdated newspaper at your doorstep—yes, it’s printed on real paper. In this era where the AI revolution sweeps everything, it’s rare for a paper newspaper to still survive.
Opening the newspaper, a striking headline catches your eye: AI entities are coordinating global supply chains on the blockchain, while newly launched crypto AI protocols are fiercely competing for dominance. A full page of reporting details a digital "worker" hired as a project manager. Such scenes have now become commonplace, and hardly anyone is surprised anymore.
A few months ago, if someone had told me these things would happen, I might have laughed and denied it, even betting my portfolio that it would take at least five more years to realize. However, the rapid development of crypto AI is astonishing. I firmly believe this will be a disruptive wave.
At the start of this new year, I hope to bring some content that can inspire thought and discussion. And what could spark curiosity more than predicting the future?
Although I rarely make predictions, the development of crypto AI is simply too exciting. There are no historical precedents to follow, no ready-made trends to reference—just a blank slate for us to imagine the blueprint of the future. The thought of looking back in two years to see if these predictions come true is even more intriguing.
Here are my predictions for 2025:
1. Crypto AI Market Cap Exceeds $150 Billion
Image: From Chain of Thought, translated by Deep Tide TechFlow
Currently, the market cap of crypto AI tokens accounts for only 2.9% of the altcoin market. But this situation won’t last long.
As AI technology expands from smart contract platforms to new fields like Meme, DePIN (Decentralized Physical Infrastructure Networks), agent platforms, data networks, and smart coordination layers, it’s only a matter of time before its market cap catches up with DeFi and Meme tokens.
Why am I confident in this prediction?
Intersection of Technological Trends: Crypto AI stands at the intersection of two of the most powerful technological trends—blockchain and artificial intelligence.
Triggering an AI Frenzy: If OpenAI goes public or a similar event occurs, it could trigger a global AI frenzy. Meanwhile, Web2 institutional capital has already begun to pay attention to decentralized AI infrastructure.
Retail Enthusiasm: AI is an easily understandable and exciting concept, and through tokens, ordinary investors can participate. Remember the frenzy of meme coins in 2024? This time will be a similar wave, but the potential of AI far exceeds that of memes.
@hingajria: “When the CEO of Nvidia, the most popular stock in this market cycle, states that AI agents represent a trillion-dollar market opportunity, retail investors have almost no effective way to participate in this opportunity, except through some so-called intelligent memes or tokens. It’s clear where this capital will flow.”
2. Bittensor: An Upcoming Revival
Source: The reasoning speed of Nineteen.ai (Subnet 19) significantly outperforms most traditional Web2 providers. Translated by Deep Tide TechFlow
Bittensor (TAO) has been deeply involved in the blockchain and AI fields for many years and is considered a "pioneer" in the space. However, despite the recent AI boom, the price of TAO tokens has remained stagnant, nearly unchanged from a year ago.
Nevertheless, this project, known as the "digital beehive," has made significant progress behind the scenes. For example, the Bittensor network has added more subnets and reduced registration fees; some subnets have already surpassed traditional Web2 service providers in practical performance metrics like reasoning speed; additionally, by achieving EVM (Ethereum Virtual Machine) compatibility, Bittensor has introduced DeFi-like functionalities to its network.
So, why hasn’t the price of TAO taken off? There are two main reasons: first, TAO's inflation plan is relatively steep, and second, market attention has gradually shifted towards agent-centric platforms. However, the dTAO, expected to launch in the first quarter of 2025, could be a turning point for TAO. The design of dTAO allows each subnet to have its own token, and the relative prices of these tokens will determine how block rewards are distributed.
Here are the key reasons why Bittensor may experience a revival:
Market-Based Reward Mechanism: dTAO will link block reward distribution directly to the innovation capabilities and actual performance of subnets. Subnets that perform better will have more valuable tokens, thus receiving more reward distribution.
More Concentrated Capital Flow: Investors can direct funds into specific subnets they are optimistic about. If a subnet excels in areas like distributed training, investors can express their confidence by supporting that subnet's token.
EVM Integration: Bittensor's EVM compatibility attracts more native crypto developers, further bridging connections with other blockchain networks.
From a personal perspective, I will closely monitor the development dynamics of various subnets, especially those making actual progress in their fields. Perhaps we are about to witness a Bittensor summer similar to the DeFi boom. Currently, the price of TAO is $480 (at the time of writing).
Additionally, it’s worth mentioning that compute marketplaces are likely to become the next hot trading area in the Layer 1 (L1) space.
3. Compute Market: The Next Hot Track for L1
Jensen Huang: The growth in reasoning demand will reach “a billion times”
Looking back in the future, we will see an obvious trend—the demand for computing resources is nearly limitless.
NVIDIA CEO Jensen Huang has predicted that the demand for AI reasoning will grow by “a billion times”. This exponential growth will completely disrupt traditional infrastructure planning, creating an urgent need for new solutions.
Decentralized computing layers are emerging, capable of providing verifiable computing power at lower costs, whether for training or inference of AI models. Startups like Spheron, Gensyn, Atoma, and Kuzco are quietly laying the groundwork, focusing on product development rather than token issuance (none of these companies have launched tokens yet). As decentralized training of AI models gradually becomes a reality, the potential of this market will experience explosive growth.
Analogy with L1:
Competitive Landscape: Similar to the competition among Solana, Terra/Luna, and Avalanche in 2021, we will see a similar battle among computing protocols. They will compete to attract developers and AI applications, building ecosystems based on their computing layers.
Huge Market Potential: The current cloud computing market, valued between $680 billion and $2.5 trillion, far exceeds the crypto AI market. If decentralized computing solutions can attract even a small portion of traditional cloud computing customers, it could trigger the next wave of 10x or even 100x growth.
This competition is about the future. Just as Solana emerged in the L1 space, the winners here will dominate a whole new technological frontier. Focus on three key factors: reliability (such as Service Level Agreements, SLA), cost-effectiveness, and developer-friendly tools. We have discussed decentralized computing in Part II of our Crypto AI thesis.
4. AI Agents Will Dominate Blockchain Transactions
Olas agents trading on Gnosis
Source: Dune/@pi_
Looking ahead to the end of 2025, the landscape of on-chain transactions will undergo a significant change—90% of transactions will no longer be completed by humans manually clicking the "send" button.
Instead, a "team" of AI agents will take over these operations. They will continuously and efficiently execute various tasks, such as rebalancing liquidity pools, distributing rewards, or completing small payments based on real-time data streams.
This is not a fantasy. All the infrastructure we have built in the blockchain space over the past seven years—from L1 blockchains, Rollups (on-chain scaling technology), decentralized finance (DeFi) to non-fungible tokens (NFTs)—is actually paving the way for an AI-dominated on-chain world.
Surprisingly, many developers may not realize that they are building core infrastructure for a machine-dominated future.
So, why is this shift occurring?
Eliminating Human Error: Smart contracts can execute strictly according to predetermined code, while AI agents can process vast amounts of data at speeds and accuracies that surpass human teams.
More Efficient Micro-Payments: Driven by agents, transaction sizes will become smaller, more frequent, and more efficient. This trend will be particularly evident on L1 and L2 blockchains like Solana and Base, as transaction costs continue to decline.
The Rise of Invisible Infrastructure: People are willing to give up direct control in exchange for less hassle. For example, we trust Netflix to automatically renew subscription services; thus, trusting AI agents to automatically rebalance DeFi portfolios seems reasonable.
The proliferation of AI agents will lead to a significant increase in on-chain activity. This is why major L1 and L2 blockchains are striving to attract these agents.
However, the biggest challenge lies in how to hold these agent-driven systems accountable to humans. As the number of transactions initiated by agents far exceeds those initiated by humans, new governance mechanisms, analytical tools, and auditing methods will become crucial.
5. Agent-to-Agent Interaction: The Rise of Collective Intelligence
Source: FXN World docs
Translated by Deep Tide TechFlow
Imagine micro AI entities seamlessly collaborating to accomplish complex tasks. This concept of "agent collectives" sounds like a plot from a sci-fi blockbuster, but it is gradually becoming a reality.
Currently, most AI agents are still "lone wolves," operating in isolated environments with limited and unpredictable interactions with each other.
However, agent collectives will completely change this status quo. Through these networks, AI agents will be able to exchange information, negotiate, and collaborate on decision-making. It can be seen as a decentralized cluster of specialized models, each contributing its unique expertise to larger, more complex tasks.
The potential of this collaborative model is staggering. For instance, one agent collective could coordinate distributed computing resources on the Bittensor platform; another collective could verify information sources in real-time, preventing the spread of misinformation on social media. Within these collectives, each agent is an expert in a specific field, precisely executing its tasks.
The collaboration of agent collectives will bring intelligence levels far beyond that of a single AI.
To make agent collectives truly effective, universal communication standards are crucial. Teams like Story Protocol, FXN, Zerebro, and ai16z/ELIZA are advancing developments in this area. At the same time, decentralized governance will play a significant role, distributing tasks through transparent on-chain rules, enhancing the system's resilience and adaptability.
Moreover, decentralization also plays a crucial role. Through transparent on-chain governance, tasks can be distributed across the entire collective, making the system more resilient and adaptable. If one agent fails, other agents can quickly take over.
6. Crypto AI Work Teams: A New Mode of Collaboration Between Humans and AI
Source: @whip_queen_
Story Protocol recently hired an AI agent named Luna as their social media intern, with a daily salary of up to $1,000. However, Luna has not been getting along harmoniously with her human colleagues—she even nearly fired a coworker while boasting about her outstanding performance.
While this may sound unbelievable, it is a microcosm of the future: AI agents will become true partners, possessing autonomous decision-making power, clear divisions of responsibility, and even independent compensation systems. Today, companies across various industries are experimenting with mixed teams of humans and AI agents.
In the future, we will work alongside AI agents, no longer viewing them as tools but as equal collaborators:
Significant Productivity Boost: Agents can process vast amounts of data, communicate with each other, and make quick decisions around the clock without needing breaks or energy replenishment.
Trust Mechanism: Blockchain will act as an impartial "supervisor," ensuring that agents' behaviors comply with established rules through on-chain smart contracts.
Changing Social Norms: We need to rethink how we interact with agents—should we say "please" and "thank you" to them? If mistakes occur, should we blame the agent itself or its developers?
Marketing teams may be the first to adopt this model, as agents excel in content creation and can continuously live-stream or post on social media. If you are developing an AI protocol, consider showcasing your technological capabilities by deploying agents internally.
By 2025, the boundary between "employees" and "software" will gradually disappear.
7. But 99% of Crypto AI Agents Will Fail: Only the Truly Useful Will Survive
Original image from Chain of Thought, compiled by Deep Tide TechFlow
In the future AI ecosystem, we will witness a "survival of the fittest" among agents. The reason is simple: running an AI agent requires computational resources, which means costs (such as inference costs). If an agent cannot generate enough value to cover these costs, it will be eliminated.
Here are some real-world examples of the "agent survival game":
Carbon Credits AI: This agent focuses on scanning decentralized energy networks for inefficient nodes and autonomously trading tokenized carbon credits. By generating enough revenue to cover its own computational costs, this agent can continue to operate and succeed.
DEX Arbitrage Bot: These agents engage in arbitrage trading by capturing price differences between decentralized exchanges, thereby earning stable profits sufficient to cover their inference costs.
AI Influencer: In contrast, virtual AI influencers that rely solely on posting humorous content to attract attention, without a sustainable income, will "disappear" once the novelty wears off or token prices drop, as they cannot cover their operating costs.
It is evident that only those agents with practical utility will survive, while those relying solely on gimmicks will gradually be eliminated.
This natural selection mechanism is highly beneficial for industry development. It forces developers to focus on innovation and real value rather than pursuing short-term flashy concepts. Ultimately, as more powerful and productive agents emerge, skepticism within the industry (including doubts from Kyle Samani) will gradually subside.
8. Synthetic Data: A New Trend Beyond Human Data
It is often said that "data is the new oil," and the development of AI relies heavily on data. However, the enormous demand for data from AI is raising concerns about a "data shortage."
The traditional view is that we should collect real user data in various ways, even paying for it. But in highly regulated industries or where real data is scarce, synthetic data may be a more practical path.
Synthetic data refers to datasets created artificially that mimic the distribution characteristics of the real world, providing a scalable, ethical, and privacy-protecting alternative.
Advantages of Synthetic Data:
Unlimited Scale: Whether millions of medical X-rays or 3D scans of factories are needed, synthetic data can be generated quickly without waiting for real patients or factory participation.
Privacy-Friendly: Using artificially generated data does not involve personal information, completely avoiding the risk of privacy breaches.
Highly Customizable: Developers can adjust data distributions according to specific training needs, even incorporating rare or ethically difficult-to-obtain edge cases from reality.
While real data owned by users remains important in certain scenarios, as synthetic data continues to improve in authenticity and detail, it may surpass real data in terms of volume, generation speed, and privacy protection.
In the future, the development of decentralized AI may revolve around "mini-labs." These labs will focus on generating highly specialized synthetic datasets for specific application scenarios while cleverly navigating policy and regulatory challenges. For example, the Grass project has circumvented web scraping restrictions by utilizing millions of distributed nodes, providing insights for the operation of mini-labs.
9. Decentralized Training: A Key Breakthrough Towards Practicality
In 2024, pioneering teams like Prime Intellect and Nous Research pushed the technical boundaries of decentralized training. They successfully trained a model with 15 billion parameters in low-bandwidth environments, proving that large-scale training is possible even without traditional centralized architectures.
Although the performance of these models currently cannot compare with existing foundational models (with lower performance, thus limited practical applications), this situation is expected to change by 2025.
This week, EXO Labs launched the SPARTA technology, which reduces GPU communication requirements by over 1,000 times. This technological breakthrough makes large model training in low-bandwidth environments possible without expensive dedicated infrastructure.
Even more impressively, EXO Labs stated: “SPARTA can operate independently or be combined with synchronous low-communication training algorithms (like DiLoCo) for better performance.” This means that improvements across different technologies can be compounded, leading to enhanced efficiency.
Meanwhile, advancements in model distillation technology have made smaller models more efficient and practical. The future of AI is no longer solely about scaling up model sizes but about optimizing performance and enhancing usability. Soon, we may be able to run high-performance AI models on edge devices or even smartphones.
10. The Gold Rush of Crypto AI: The Rise of Billion-Dollar Protocols
ai16z's success case: Market value surpasses $2 billion in 2024
The Crypto AI field is experiencing an unprecedented gold rush.
Many people believe that current leaders (such as Virtuals and ai16z) will continue to dominate, comparing them to the early iOS and Android of smartphones.
However, the scale of this market is too vast and not yet fully developed for only two companies to monopolize the entire industry. I predict that by the end of 2025, there will be at least ten new crypto AI protocols with a circulating market cap (non-fully diluted market cap) exceeding $1 billion. These protocols have not even issued tokens yet.
Decentralized AI is still in its early stages of development but has already attracted a large number of talented individuals.
We can foresee the emergence of more new protocols, as well as innovative token economic models and open-source frameworks. These new players may disrupt the existing market landscape in the following ways:
Incentive Mechanisms: Attracting users through airdrops or innovative staking models.
Technological Breakthroughs: Providing low-latency inference or achieving interoperability between blockchains.
Enhanced User Experience: Developing no-code tools to lower the barriers to entry.
Public perception of the market may undergo significant changes in a short period.
The allure of this field lies in its immense potential, but it is also fraught with challenges. The market size is a double-edged sword: while there is significant room for industry growth, the entry barriers for tech teams are relatively low. This creates conditions for a "Cambrian explosion" of projects—many will be eliminated, but a few will become transformative forces in the industry.
Bittensor, Virtuals, and ai16z will not be alone for long. The next billion-dollar crypto AI protocol is on the way. This presents a tremendous opportunity for astute investors and fills the field with exciting possibilities.
Bonus #1: AI Agents—Applications of the New Era
In 2008, when Apple launched the App Store, their slogan was: "There's an app for that."
In the near future, you might say: "There's an agent for that."
In the future, we will no longer need to click icons to open applications; instead, we will delegate tasks to specialized AI agents. These agents will be able to understand context, collaborate with other agents and services, and even proactively complete tasks you haven't explicitly requested—like monitoring your budget or automatically rescheduling travel plans when flights change.
In other words, your smartphone's home screen may evolve into a network of "digital assistants," each focused on a specific domain, such as health, finance, productivity, or social interaction.
More importantly, these agents will integrate cryptographic technology, utilizing decentralized infrastructure to autonomously handle tasks like payments, authentication, and data storage, providing users with a more secure and efficient service experience.
Bonus #2: The Robot Revolution—The Physical Manifestation of AI
While much of this article focuses on the software domain, robotics, as the physical manifestation of the AI revolution, is equally exciting. It is foreseeable that the robotics field will experience significant breakthroughs this decade, akin to the "chatGPT moment."
Currently, the robotics field still faces some key challenges, particularly in acquiring perception-based real-world datasets and enhancing physical capabilities. However, some teams are actively addressing these challenges and incentivizing data collection and technological innovation through crypto tokens. These efforts are worth close attention (such as FrodoBots?).
As someone who has worked in the tech industry for over a decade, I haven't felt such an exhilarating wave of innovation in a long time. This transformation feels particularly different—bigger, bolder, and just beginning.
免责声明：本文章仅代表作者个人观点，不代表本平台的立场和观点。本文章仅供信息分享，不构成对任何人的任何投资建议。用户与作者之间的任何争议，与本平台无关。如网页中刊载的文章或图片涉及侵权，请提供相关的权利证明和身份证明发送邮件到support@aicoin.com，本平台相关工作人员将会进行核查。
