GrayScale: How can Crypto accelerate the arrival of the AI era?

Original Title: “AI Is Coming - Crypto Can Help Make It Right”

Author: Grayscale Research

Compiled by: Felix, PANews

Grayscale announced yesterday the launch of a new fund focused on decentralized AI, Grayscale Decentralized AI Fund LLC. The fund's initial projects include Bittensor (TAO), Filecoin (FIL), Livepeer (LPT), Near (NEAR), and Render (RNDR), with Near, Filecoin, and Render being the highest-weighted assets in the fund. Following this news, related tokens surged significantly. Subsequently, Grayscale published an article interpreting AI and decentralized AI, explaining the reasons for its emphasis on them. Below is the full text.

Artificial Intelligence (AI) is one of the most promising emerging technologies of this century, with the potential to exponentially enhance human productivity and drive breakthroughs in medicine. While AI may be important today, its influence will only grow. According to PwC, by 2030, AI is expected to grow into a $15 trillion industry.

However, this promising technology also faces challenges. As AI technology becomes more powerful, the power in the AI industry is concentrated in the hands of a few companies, posing potential risks to society. This has raised serious concerns about deepfakes, embedded biases, and data privacy risks. Fortunately, cryptographic technology offers potential solutions to some of these issues through its characteristics of decentralization and transparency.

This article will explore the problems brought by centralization and how decentralized AI can help address some of these drawbacks. It will also discuss the intersection of Crypto and AI, highlighting crypto applications in this field that have shown early signs of adoption.

Problems of Centralized AI

The current development of AI faces certain risks and challenges. The network effects and intense capital requirements of AI are so significant that many AI developers outside of large tech companies, such as small firms or academic researchers, either struggle to access the resources needed for AI development or cannot monetize their work. This limits overall competition and innovation in AI.

As a result, the influence over this critical technology is primarily concentrated in a few companies like OpenAI and Google, raising serious questions about AI governance. For example, in February of this year, Google's AI image generator Gemini was reported to have racial biases and historical inaccuracies, allegedly manipulating its model. Additionally, in November of last year, a six-member board decided to dismiss OpenAI CEO Sam Altman, exposing the fact that a small number of people control the companies developing these models.

As the influence and importance of AI grow, many are concerned that a single company could hold decision-making power over AI models that have a significant impact on society. There is a risk that barriers could be set to shut others out or that models could be manipulated for personal gain at the expense of others.

How Decentralized AI Can Help

Decentralized AI refers to AI services that utilize blockchain technology to distribute ownership and governance of AI in a way that enhances transparency and accessibility. Grayscale Research believes that decentralized AI has the potential to liberate these important decisions from closed environments for public ownership.

Blockchain technology can help developers increase access to AI and lower the barriers for independent developers to build and monetize their work. This will help enhance overall AI innovation and competition, balancing against models developed by tech giants.

Moreover, decentralized AI can help democratize investment in AI. Currently, there are few ways to gain returns related to AI development, aside from investing in some tech stocks. Meanwhile, substantial private capital has been allocated to AI startups and private companies (with $47 billion in 2022 and $42 billion in 2023). Thus, only a small number of venture capitalists and accredited investors can benefit from 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 of Fields Developing Today?

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 return rate in the AI sector of crypto assets is 20%, outperforming the vast majority of crypto tracks. Additionally, according to Kaito data, the AI sector currently has the highest "narrative mind share" on social platforms compared to other sectors like DeFi, Layer2, Meme, and RWA (with the highest market attention).

Recently, some prominent figures have begun to embrace this emerging field, aiming to address the flaws of centralized AI. In March of this year, Emad Mostaque, founder of AI company Stability AI, left the company to pursue decentralized AI, stating, "It’s time to ensure that AI remains open and decentralized." Additionally, Erik Vorhees, founder of ShapeShift, recently launched Venice.ai, a privacy-focused AI service with end-to-end encryption capabilities.

Figure 1: The performance of AI Universe has nearly surpassed 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 Needed 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 address AI-related issues, such as the rise of bots and deepfakes, and model verification (e.g., WLD, TRAC, NUM)

Figure 2: AI and Crypto Market Map

Source: Grayscale Investments. The protocols included are illustrative examples.

Networks Providing Infrastructure for AI Development

The first category includes networks that provide permissionless open architectures built specifically for the overall development of AI services. These assets do not focus on a specific AI product or service but rather on creating underlying infrastructure and incentive mechanisms for various AI applications.

NEAR stands out in this category, with its founder being a co-creator of the "Transformer" architecture that powers AI systems like ChatGPT. In May of this year, NEAR announced its focus on building a user-owned AI ecosystem, committed to optimizing user privacy and sovereignty. In late June, NEAR launched its AI incubator program to develop NEAR-native foundational models, data platforms for AI applications, AI agent frameworks, and computing markets.

Bittensor is a platform that uses the TAO token to economically incentivize AI development. 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 permissionless APIs to build specific AI applications by querying miners in the Bittensor subnets.

This category also includes other protocols like Fetch.ai and Allora Network. Fetch.ai is a platform for developers to create complex AI assistants (i.e., "AI agents") and recently merged with AGIX and OCEAN, with a total market cap of about $7.5 billion. Another is Allora Network, a platform focused on applying AI to financial applications, including automated trading strategies for DEX and prediction markets. Allora has not yet issued a token and conducted a round of strategic financing in June, raising a total of $35 million in private funding.

Resources Needed for AI Development

The second category includes assets that provide the necessary resources for AI development in the form of computing, storage, or data.

The rise of AI has created massive demand for computing resources in the form of GPUs. Decentralized GPU markets like Render (RNDR), Akash (AKT), and Livepeer (LPT) provide idle GPU supply for developers training models, performing model inference, or rendering 3D generative AI. It is estimated that Render offers around 10,000 GPUs, primarily targeting artists and generative AI; while Akash provides 400 GPUs, focusing on AI developers and researchers. Meanwhile, Livepeer recently announced a new AI subnet initiative aimed at executing AI inference tasks such as text-to-image, text-to-video, and image-to-video by August 2024.

In addition to requiring substantial computing resources, AI models also need vast amounts of data. Therefore, the demand for data storage has surged. Data storage solutions like Filecoin (FIL) and Arweave (AR) can serve as decentralized and secure 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 breaches.

Finally, existing AI services, such as OpenAI and Gemini, can continuously access real-time data through Bing and Google Search. This puts all other AI model developers, aside from tech companies, at a disadvantage. However, data scraping services like Grass and Masa can help create a level playing field by allowing individuals to profit from providing application data for AI model training while maintaining control and privacy over their personal data.

Assets Attempting to Solve AI-Related Issues

The third category includes assets that attempt to address AI-related issues, including the rise of bots, deepfakes, and content provenance.

Another significant issue with AI is the proliferation of bots and misinformation. AI-generated deepfakes have already impacted presidential elections in India and Europe, and experts are "very concerned" about the upcoming U.S. presidential campaign, fearing a massive wave of "false information" driven by deepfakes. Assets aimed at helping to address issues related to deepfakes by establishing verifiable content provenance include Origin Trail (TRAC), Numbers Protocol (NUM), and Story Protocol. Additionally, Worldcoin (WLD) seeks to address the bot problem through unique biometric identification.

Another risk of AI is ensuring trust in the models themselves. How can one trust that the AI results received have not been tampered with or manipulated? Currently, several protocols help address this issue through cryptography, zero-knowledge proofs, and fully homomorphic encryption (FHE), such as Modulus Labs and Zama.

Conclusion

While these decentralized AI assets have achieved initial results, they are still in the early stages. Earlier this year, venture capitalist Fred Wilson stated 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 crypto use cases related to AI will become increasingly important, with the potential for these two rapidly evolving technologies to mutually benefit each other.

Many signs indicate that the AI era is approaching, which will have profound impacts, both positive and negative. By leveraging the characteristics of blockchain technology, it is believed that Crypto can ultimately help mitigate some of the dangers posed by AI.