AI and encryption: The fusion of emerging technologies and the future

In today's technology sector, artificial intelligence and encryption are two of the most prominent fields. Both industries are filled with innovation and potential, achieving success in one of these fields is already remarkable, let alone making contributions in both.

Some entrepreneurs are trying to combine AI and encryption technology to create new business models. One notable direction is to build AI infrastructure using the principles of Web3. With the rise of large language models, the demand for AI infrastructure has surged. AI systems require strong computing power support and consume a lot of electricity. The traditional centralized data center model has a high threshold, while the decentralized concept of Web3 may lower the barriers to participation, allowing more people to get involved.

AI+DePIN: Decentralized Physical Infrastructure

DePIN(Decentralized Physical Infrastructure Networks) refers to decentralized physical infrastructure networks. It includes three core elements:

1. Decentralization: Unlike the centralized model of traditional cloud service providers, DePIN allows anyone to contribute resources.

2. Physical infrastructure: Services are provided by physical hardware, such as servers, IoT devices, etc.

3. Token Economy: Incentivize community participation in building the network through tokens.

Bitcoin mining can be seen as the most successful DePIN case. Now some projects are attempting to apply the DePIN concept to the AI field.

! [Bitcoin is sideways, the cottage is basically cool, and the next thing to rely on AI computing power to mine? ] ](https://img-cdn.gateio.im/webp-social/moments-0acdc52fb50c61fd46c6f67e4c0009ef.webp)

Classification of AI Demand for Computing Power

AI computation can be roughly divided into three categories:

1. Large model training: highest requirements, needs top-tier GPUs.

2. Large model inference: requirements are secondary, multiple companies can provide.

3. Edge Small Model: Lower requirements, ordinary devices can also participate.

Ordinary users' devices can participate in the latter two types of computing tasks.

AI+DePIN Representative Projects

IO.NET

IO.NET is a decentralized AI computing power platform based on Solana, valued at 1 billion USD. It aggregates idle GPU resources from independent data centers, encryption miners, and is mainly aimed at AI startups and developers.

Aethir

Aethir is a cloud computing platform that allows users to share GPU resources. It has partnered with large game studios and has a large user base. Aethir recently launched the "Inspector Node" public sale.

Grass

Grass allows users to earn points by providing idle network bandwidth. It has completed financing led by a well-known investment firm.

EMC (EdgeMatrix Computing)

EMC is an edge computing protocol that allows users to participate in AI services using regular computers and earn rewards. It has completed financing and is listed on some exchanges.

Conclusion

The combination of AI and encryption technology may create new opportunities. The DePIN model might allow more people to participate in the construction of AI infrastructure. In the future, our everyday devices may contribute computing power to AI systems while idle, thereby creating value for us. This integration represents a new direction in technological development.