Top 10 Predictions for the Encryption AI Industry in 2025

With the rapid development of the AI industry, the encryption AI field has also seen rapid growth. A researcher focused on encryption AI has made 10 predictions for 2025. The following are the main points of the predictions:

1. The total market value of encryption AI tokens reaches 150 billion USD

Currently, the market value of encryption AI tokens accounts for only 2.9% of the market value of altcoins, but this proportion is expected to increase significantly. AI encompasses various aspects from smart contract platforms to meme, DePIN, Agent platforms, data networks, and intelligent coordination layers, and its market position is expected to be on par with DeFi and meme.

Encryption AI is at the intersection of the two most powerful technologies and may trigger a global frenzy for AI. Web2 capital has begun to focus on decentralized AI infrastructure. The concept of AI is easy to understand and exciting, allowing retail investors to invest through tokens, with hopes of sparking a gold rush similar to the 2024 meme.

2. Bittensor Revival

As a veteran project in the encryption AI field, the decentralized AI infrastructure Bittensor(TAO) has been running for many years. Despite the AI craze, its token price has remained at the level it was a year ago.

Bittensor's Digital Hivemind ( has quietly achieved a leap: the registration fees for more subnets are lower, the performance of subnets in practical metrics such as inference speed is better than Web2 counterparts, and EVM compatibility will bring DeFi-like functionalities to the Bittensor network.

dTAO) is expected to launch in the first quarter of 2025, which could be a significant turning point. Each subnet will have its own token, and the relative prices of these tokens will determine how emissions are allocated. Market-based emissions will directly link block rewards to innovation and actual performance. Investors can target specific subnets they are optimistic about. EVM compatibility will attract a broader community of encryption-native developers.

3. The calculation market is the next "L1 market"

The demand for computing is growing exponentially, disrupting traditional infrastructure plans and urgently requiring new solutions. The decentralized computing layer provides raw computing ( for training and inference ) in a verifiable and cost-effective manner. Some startups are building a solid foundation, focusing on products rather than tokens.

Similar to the competition in L1 in 2021, there will also be competition among computing protocols to attract developers and AI applications. If these decentralized computing solutions can attract some traditional cloud customers, they could see growth of 10x or even 100x. The winners will dominate this new field, and it is worth paying attention to their reliability, cost-effectiveness, and developer-friendliness.

4. AI agents will flood blockchain transactions

By the end of 2025, 90% of on-chain transactions will be executed by AI agents, which continuously rebalance liquidity pools, allocate rewards, or execute small payments based on real-time data feedback. The L1, rollup, DeFi, NFT, and other infrastructures built over the past seven years have paved the way for AI to operate on-chain.

AI agents can process large amounts of data faster and more accurately than humans, reducing human errors. Transactions will become smaller, more frequent, and more efficient. Humans are also willing to relinquish direct control to reduce hassle. All L1/L2 are embracing agents.

The biggest challenge is to make these agent-driven systems accountable to humans. As the proportion of transactions initiated by agents continues to grow, new governance mechanisms, analytical platforms, and auditing tools will be needed.

5. The Rise of Intelligent Body Clusters

The concept of Agent clusters refers to the seamless collaboration of micro AI agents executing grand plans. Current AI agents are mostly "lone wolves", with minimal interaction and unpredictability. Agent clusters will change this situation, allowing AI agent networks to exchange information, negotiate, and make collaborative decisions.

These cluster networks will generate more powerful intelligence than a single isolated AI. Universal communication standards are crucial, and Agents need to be able to discover, verify, and collaborate. Some teams are laying the groundwork for the emergence of Agent clusters.

Decentralization plays a key role in this. Under the management of transparent on-chain rules, tasks are assigned to various clusters, making the system more resilient and adaptable.

6. The encryption AI working team will be a human-machine hybrid.

In the future, AI Agents will become true collaborators, possessing autonomy, responsibility, and even salaries. Companies across various industries are conducting beta tests on human-machine hybrid teams. Collaborating with AI Agents will enhance productivity, establish trust through smart contracts, and social norms will continue to evolve. The boundary between "employees" and "software" will begin to disappear in 2025.

7. 99% of AI Agents Will Vanish

In the future, we will see a "Darwinian" elimination among AI agents. Running AI agents requires expenditures in the form of computing power ( and reasoning costs ). If an Agent cannot generate enough value to pay its "rent," it will be eliminated.

Utility-driven agents will thrive, while distraction-driven agents will gradually become irrelevant. This elimination mechanism is beneficial for industry development, forcing developers to innovate and prioritize production use cases over gimmicks.

8. Synthetic data exceeds human data

Synthetic data is a dataset generated artificially, designed to mimic the data distribution of the real world. It provides a scalable, ethical, and privacy-friendly alternative to human data. Synthetic data has advantages such as unlimited scalability, privacy-friendliness, and customizability.

The next wave of decentralized AI may center around "micro-laboratories," which can create highly specialized synthetic datasets tailored for specific use cases. These micro-laboratories will cleverly circumvent policy and regulatory barriers in data generation.

9. Decentralized training is more useful

Some pioneers have made breakthroughs in decentralized training. Although the performance of these models is currently low, it is expected that this will change by 2025. New technologies can significantly reduce communication between GPUs, allowing large model training over slow bandwidth without the need for specialized infrastructure.

With technological advancements, micro-models will become more practical and efficient. The future of AI does not lie in scale, but in becoming better and easier to use. High-performance models that can run on edge devices and even mobile phones are expected to emerge soon.

10. Ten new encryption AI protocols have a circulating market value of 1 billion USD.

By the end of 2025, it is expected that at least ten new encryption AI protocols ( that have not yet launched tokens ) will have a circulating market value exceeding $1 billion. Decentralized AI is still in its infancy, with a growing talent pool. New protocols may replace existing participants through incentives, technological breakthroughs, and improvements in user experience.

The market size is enormous, but the entry barrier for technology teams is relatively low. This lays the foundation for the explosive growth of projects; many projects will gradually disappear, but a few will possess transformative power. Existing leading projects will not always remain at the forefront; the next $1 billion encryption AI protocol is about to emerge.