作者：Grayscale Research
Compiled by: Felix, PANews
Grayscale announced yesterday the launch of Grayscale Decentralized AI Fund LLC, a new fund focused on decentralized AI. The first batch of projects in the fund include Bittensor (TAO), Filecoin (FIL), Livepeer (LPT), Near (NEAR) and Render (RNDR), among which Near, Filecoin and Render are the highest-weighted assets in the fund. Influenced by this news, the relevant tokens rose sharply. Subsequently, Grayscale published an article to interpret AI and decentralized AI, and explained the reasons for its importance. The following is the full text.
Artificial intelligence (AI) is one of the most promising emerging technologies of this century, with the potential to exponentially increase human productivity and power medical breakthroughs. While AI may be important today, its impact will only grow. PwC estimates that by 2030, AI will grow into a $15 trillion industry.
However, this promising technology also faces challenges. As AI technology becomes more powerful, power in the AI industry is concentrated in a few companies, which is potentially harmful to society. This has also raised serious concerns about deep fakes, embedded biases, and data privacy risks. Fortunately, crypto technology offers potential solutions to some of these problems with its decentralized and transparent nature.
This article will explore the problems caused by centralization and how decentralized AI can help solve some of these drawbacks. It will also discuss the intersection of Crypto and AI, focusing on crypto applications in the field that have shown early signs of adoption.
The Problem with Centralized AI
The current development of AI faces certain risks and challenges. The network effects and intensive capital requirements of AI are so significant that many AI developers outside of large tech companies, such as small companies or academic researchers, either struggle to obtain the resources needed for AI development or are unable to monetize their work. This limits overall competition and innovation in AI.
As a result, the influence on this key technology is mainly concentrated in the hands of a few companies such as OpenAI and Google, raising serious questions about AI governance. For example, in February this year, Google's AI image generator Gemini was exposed for racial discrimination and historical errors, suspected of manipulating the model. In addition, last November, a six-member board of directors decided to fire OpenAI CEO Sam Altman, exposing the fact that a few people control the companies that develop these models.
As AI becomes more influential and important, many people worry that a single company could seize the power to make decisions about AI models that have a huge impact on society. They could even build walls around AI or manipulate models for their own benefit at the expense of others.
How Decentralized AI Can Help
Decentralized AI refers to AI services that use blockchain technology to distribute AI ownership and governance in a way that increases transparency and accessibility. Grayscale Research believes that decentralized AI has the potential to free these important decisions from closed environments and make them publicly owned.
Blockchain technology can help developers increase access to AI and lower the barrier for independent developers to build and monetize their work. This will help increase overall AI innovation and competition and maintain a balance with models developed by tech giants.
Additionally, decentralized AI can help democratize AI investing. Currently, there are few ways to access the gains associated with AI development, other than through a few tech stocks. At the same time, a large amount of private capital has been allocated to AI startups and private companies ($47 billion in 2022 and $42 billion in 2023). As a result, only a small number of VCs and accredited investors can access the gains of these companies. In contrast, decentralized AI crypto assets are open to everyone, allowing anyone to participate in the future of AI.
How is the intersection field developing now?
The intersection of Crypto and AI is still in its early stages in terms of maturity, but the market response is encouraging. As of May 2024, the AI field of crypto assets has a return rate of 20%, which is better than the vast majority of crypto tracks. In addition, according to Kaito data, compared with other tracks such as DeFi, Layer2, Meme and RWA, the AI track currently has the highest "narrative mind share" on social platforms (highest market attention).
Recently, some well-known figures have begun to embrace this emerging field and are committed to solving the shortcomings of centralized AI. In March of this year, Emad Mostaque, founder of AI company Stability AI, left the company to pursue decentralized AI, saying that "now is the time to ensure that AI remains open and decentralized." In addition, ShapeShift founder Erik Vorhees recently launched Venice.ai, a privacy-focused AI service with end-to-end encryption.
Figure 1: AI Universe has outperformed almost all crypto tracks so far this year
The intersection of Crypto and AI can be divided into three main subcategories:
Infrastructure layer: Networks that provide platforms for AI development (e.g. NEAR, TAO, FET)
Resources required for AI: Assets that provide key resources (computing, storage, data) required for AI development (e.g. RNDR, AKT, LPT, FIL, AR, MASA)
Solving AI problems: Assets that attempt to solve AI-related problems, such as the rise of bots and deepfakes, and model validation (e.g. WLD, TRAC, NUM)
Figure 2: AI and Crypto Market Map
Source: Grayscale Investments. Protocols included are illustrative examples
A network that provides infrastructure for AI development
The first category is a network that provides an open architecture without permission and is built for the overall development of AI services. These assets are not focused on a single AI product or service, but rather focus on creating the underlying infrastructure and incentives for a variety of AI applications.
NEAR stands out in this category, with its founders being the co-founders of the "Transformer" architecture, which powers AI systems such as ChatGPT. In May, NEAR announced that it would focus on building a user-owned AI ecosystem dedicated to optimizing user privacy and sovereignty. In late June, NEAR launched its AI Incubator program to develop NEAR native base models, data platforms for AI applications, AI agent frameworks, and computing markets.
Bittensor is a platform that uses TAO tokens to economically encourage the development of AI. Bittensor serves as the underlying platform for 38 subnets, each with different use cases such as chatbots, image generation, financial forecasting, language translation, model training, storage, and computing. The Bittensor network rewards the best performing miners and validators in each subnet with TAO tokens, and provides developers with a permissionless API to build specific AI applications by querying miners in the Bittensor subnet.
This category also includes other protocols such as Fetch.ai and Allora Network. Fetch.ai is a platform for developers to create complex AI assistants (i.e. "AI agents"), which recently merged with AGIX and OCEAN, with a total market value of approximately $7.5 billion. Another is the Allora Network, a platform focused on applying AI to financial applications, including DEX and automatic trading strategies for prediction markets. Allora has not yet issued a token and conducted a strategic round of financing in June, with a total private financing of $35 million.
Resources needed for AI development
The second category includes assets that provide the resources needed for AI development in the form of compute, storage, or data.
The rise of AI has created a massive demand for computing resources in the form of GPUs. Decentralized GPU marketplaces such as Render (RNDR), Akash (AKT), and Livepeer (LPT) provide a supply of idle GPUs to developers of model training, model inference, or rendering 3D generative AI. It is estimated that Render provides about 10,000 GPUs, focusing on artists and generative AI; while Akash provides 400 GPUs, focusing on AI developers and researchers. Meanwhile, Livepeer recently announced a new AI subnet plan with the goal of performing AI reasoning tasks such as text-to-image, text-to-video, and image-to-video by August 2024.
In addition to requiring a large amount of computing resources, AI models also require a large amount of data. Therefore, the demand for data storage has increased significantly. Data storage solutions such as Filecoin (FIL) and Arweave (AR) can serve as decentralized and secure network alternatives to storing AI data on centralized AWS servers. These solutions not only provide cost-effective and scalable storage, but also enhance data security and integrity by eliminating single points of failure and reducing the risk of data leakage.
Finally, existing AI services, such as OpenAI and Gemini, provide continuous access to real-time data through Bing and Google Search, respectively. This puts all other AI model developers, except for tech companies, at a disadvantage. However, data scraping services like Grass and Masa can help level the playing field by allowing individuals to profit by providing application data for AI model training while maintaining control and privacy over their personal data.
Assets that attempt to solve AI-related problems
The third category includes assets that attempt to address AI-related issues, including the rise of bots, deepfakes, and content provenance.
Another significant problem with AI is the proliferation of bots and misinformation. AI-generated deepfakes have already had an impact on presidential elections in India and Europe, and experts are "very scared" of the upcoming U.S. presidential race being flooded with "disinformation" heavily driven by deepfakes. Assets designed to help solve problems related to deepfakes by establishing verifiable sources of content include Origin Trail (TRAC), Numbers Protocol (NUM), and Story Protocol. Furthermore, Worldcoin (WLD) attempts to solve the bot problem by verifying it with a unique biometric identifier.
Another risk of AI is ensuring trust in the model itself. How can you trust that the AI results you receive have not been tampered with or manipulated? Currently, there are several protocols that help solve this problem through cryptography, zero-knowledge proofs, and fully homomorphic encryption (FHE), such as Modulus Labs and Zama.
in conclusion
Although these decentralized AI assets have achieved initial results, they are still in the early stages. Earlier this year, venture capitalist Fred Wilson said that AI and Crypto are "two sides of the same coin" and "Web3 will help us trust AI." As the AI industry continues to mature, Grayscale Research believes that these AI-related encryption use cases will become increasingly important, and these two rapidly developing technologies have the potential to complement each other.
There are many signs that the AI era is coming and will have far-reaching, positive or negative effects. By leveraging the characteristics of blockchain technology, I believe Crypto can eventually help mitigate some of the dangers of AI.
Related reading: Why VCs are betting big on Crypto x AI
