Bankless: Can Crypto’s AI protocol beat Google?

Original title: Can Crypto's AI Protocols Beat Google?

Original author: David C

Original source: bankless

Compiled by: Kate, Mars Finance

Investors are betting on decentralized virtual infrastructure. Does it have a chance?

We've been through several AI hype cycles over the past few decades, but this one feels different. The technology was there and investors saw the opportunity. But despite the huge potential for value creation, the tech giants appear poised to eat up the opportunity.

But don’t worry, Crypto’s developers are working on a solution.

As the application of artificial intelligence accelerates, the demand for computing power and data storage is also facing severe pressure. A new class of protocols is driving the development of so-called decentralized virtual infrastructure (DeVIN). The technology uses blockchain to create markets and expand access to GPUs and data storage.

In this article, we’ll explore the current state of the computational bottleneck crisis, the overall benefits of DeVIN, and the outstanding projects leading the way in this space. Let’s get started

The centralization problem of artificial intelligence

OpenAI's Sam Altman was in the news recently when he said trillions of dollars are needed to address the near-term development challenges of artificial intelligence. In fact, model training costs are expected to reach 1% of U.S. GDP (over $100 billion) in the next few years.

This historic demand is crippling computing supply lines, and the DeVIN protocol offers a potential solution to these challenges.

DeVIN is mainly divided into several categories of solutions:

• Protocol for decentralized computing

• Protocol for decentralized data storage

While the market capitalization of AI-related tokens is soaring, the reality is that many of these projects are still in their infancy. While many investor actions can be attributed to betting on the AI ​​+ Crypto Meme, these protocols are building a vision that everyone in crypto should be able to participate in - greater decentralization and greater decentralization in our AI future. More community ownership.

Let’s take a look at some of the boundary-pushing protocols in DeVIN.

• Protocol for decentralized computing

By leveraging a global network of underutilized resources, DeVIN platforms such as io.net, Gensyn, and Grass aim to solve the computing crisis of artificial intelligence.

By developing marketplaces where anyone can contribute CPU/GPU or storage power and receive rewards, these protocols not only create more competition among infrastructure providers, but also make AI development itself more competitive by democratizing access to resources. .

io.net

io.net provides access to computing power by creating a decentralized network that leverages underutilized GPUs from sources such as data centers and crypto miners, weaving them into a cohesive system that leverages traditional Delivers powerful processing power at a fraction of the cost, reducing costs and increasing the availability of cloud computing resources. Recently, io.net launched Project Ignition to open its platform to anyone who wants to provide GPU capabilities to the network ahead of the token’s launch in April.

• Revision

Gensyn also hopes to address current computing limitations with DeVIN. Like io.net, Gensyn directly connects the world's unused processing power with machine learning projects that need it, using a market-based mechanism to facilitate this exchange. Gensyn’s rigorous focus on machine learning applications sets it apart from other DeVIN projects. In addition to providing calculations, Gensyn can also verify that work and model training were completed correctly, ensuring that the training performed on someone's behalf in the network is trustworthy and reliable.

Grass

Grass is a decentralized network that collects public network data to train artificial intelligence models. By installing the browser extension, users run a node on Grass and provide available internet bandwidth to the network, allowing its partners to collect public website data and train their models. Users will then be compensated. According to data provided by users, Grass is currently in beta.

️Decentralized data storage protocol

Now, let’s move from discussing decentralized computing to looking at DeVIN for data storage.

These solutions are more mature than computing solutions, but are still in their early stages and require a decentralized data storage market to keep up with the huge resource constraints of effectively training AI. It is estimated that by 2025, more than 175 zettabytes (175T gigabytes) of data will be stored globally, reducing costs and making it more efficient to structure will have profound consequences for data training in artificial intelligence and data management Influence.

️Filecoin

Filecoin's data storage solutions impact the world's data, turning idle storage space into an active and accessible market. The protocol is known for its method of storing, encrypting, copying, and distributing data across the network, making it highly resistant to censorship or data loss and far less expensive. Filecoin provides a powerful and decentralized backbone for the data layer of artificial intelligence, which is fully consistent with the use case of blockchain as a tamper-proof data center.

The platform uses a unique model to verify storage, ensuring that data stored on the network remains secure and accurate over time. This verification process is a crucial step as Filecoin aims to prove that decentralized storage can outperform traditional cloud storage solutions.

️Arweave

Arweave approaches data storage in a novel way, positioning itself as a permanent archive for the digital age rather than a temporary repository. Rather than using a “rent” model for data storage, the protocol uses a “buy” model, offering a one-time payment to secure centuries of data storage on its decentralized network, powered by AR, a mineable of tokens) protection. Arweave’s ambitious vision to build a decentralized, permanent network is a major leap forward in the search for sustainable and secure data storage solutions.

Arweave has expanded on the data storage aspect and recently entered the world of decentralized computing with the launch of Arweave AO’s public testnet. This addition transforms Arweave from a storage solution into a holistic computing platform with a "hyper-parallel computer" capable of handling heavy computing workloads.

our direction

Although DeVIN is at the forefront of encryption technology, the road ahead is still filled with many challenges.

The above protocols have been around earlier than their market capitalizations might suggest, and as generative AI advances and AI’s biggest players begin counting down to AGI, time is of the essence.

Projects such as io.net, Gensyn, Grass, Filecoin and Arweave highlight the critical role of decentralized computing and storage in broadening the market and reducing entry costs. They also highlight the urgent need for DeVIN to address current adoption barriers to fully support AI’s growing computing needs.

While the combination of cryptocurrency and AI may seem like a meme to many, the democratization of the infrastructure needed to compete in the AI ​​space is not.

The computing and data storage challenges faced by artificial intelligence highlight the clear need for DeVIN solutions. However, fully realizing these benefits will require addressing early adoption issues, regulatory hurdles and the ongoing GPU arms race. Obstacles like these still prevent DeVIN from becoming a backbone of AI infrastructure.

There's a lot at stake for DeVIN players if they don't find product/market fit soon, and the next few months provide a unique window for decentralized computing providers to secure their presence in the market. status. The stakes are high!

标签：ai crypto AI