The combination of AI and Web3 unlocks significant potential for use cases that have not been possible with traditional technologies.

In this blog post, I outline:

1. What is Web3?
2. Why is Web3 important?
3. What are some issues with web3?
4. Some distinct use cases for AI in the Web3 ecosystem
5. What about web3 results in more efficient AI?

What is Web3?

Today, if you need to access a service, you must ask centralized corporations such as retailers, banks, and social media companies for an account. Then you can access the services they provide using this account. They provide the service, and they control your data. They also control your account and your access.

In contrast, in web3, consider that you already have an account on the “internet”. You now choose which company (e.g. bank, retailer, social media company etc.) you want to access your account through. Businesses would compete to act as agents so they can perform their services for you. Your data and your account is yours. No corporation can take it way from you.

So you now have an account on the internet, versus you having an account with one of the companies who manage your data on their own private database. So you can be confident that your account cannot be shut down by any of the corporate bad actors. It belongs to you, not the corporations.

To illustrate this point consider these 2 examples:

1. Today you share videos with people by putting them on YouTube or Vimeo or any other service. These companies host your videos. In web3, you would put your videos on the internet and YouTube would only be one of the apps with access to them. If you decide to, any other app such as Vimeo would have it too. The likes and followers would also be on the internet in addition to being on the company’s app. A version of this already exists in the LBRY project.
2. Consider you having a banking account on the internet – its not with Bank of America or Chase. Your balances, transactions, who you transact with, etc. would not be controlled solely by the banks. Instead, what they would do is to provide you services such as making payments to others or helping you get loans. The accounts and transactions themselves would be on the internet.  No one bank would control your assets and money. Bitcoin, Ethereum, or another cryptocurrency is an example of that.

Why Web3?

The underlying idea is that individual users would have more control in a web3 world rather than solely being customers of corporations. Each corporation would compete with others for your business just like they do today.

In addition, corporations would have an easier time too.  They won’t have to spend money on technology and data processing that is a bulk of the spending today. They can instead spend money on what truly differentiates them in the market. The underlying plumbing of data processing and storage would be handled by the web3 protocol layer. Every company would not need to build up the same infrastructure over and over again.

Processing of transactions would have lower latencies and will be more accurate too. Today a large proportion of operations and IT budget is spent to ensure that databases across various companies are in sync. Updates to one system reflect on another system after 1, 2 or even 3 days. In contrast, the underlying web3 protocol can be seen as a giant database on the internet. Everyone will always have the latest information on their fingertips if they have been given access. A popular use case is payments which would be near instantaneous. No longer would you have to wait a few hours or days to see your payment accounts reflect the latest.

The notion of web3 is also appealing because of societal risk. With the increasing digitization of the world, there is a real risk of being excluded based on political and other social decisions. We have seen that with actions by the Canadian government when they froze bank accounts of protesting truckers, GoFundMe as a private company withholding contributions from people to causes they deemed unacceptable, and even companies such as AWS taking servers of companies offline stating that the targeted company was against currently acceptable political and social beliefs. Since there are always two sides to any discussion, the concept of web3 appeals to people who want to promote individual choice over group think.

The Concerns with Web3

While web3 will alleviate some of these concerns, there are of course significant tradeoffs too. Today the world of web3 is powered by blockchain technology and has struggled to solve a few thorny issues.

1. Credentials: Users have to be more tech savvy in maintaining their credentials. Today if you forget your password you can request a new one. In a web3 world, that’s not going to be as straightforward. There are still hybrid models available where corporations will handle that for you, but then your data is not truly going to be your own.
2. Privacy: There is also the question of privacy. Today, because of the way web3 works, user transactions are kept on the internet. Thus they are theoretically visible to everyone else. Even though much of this information is encrypted and often split up and stored in multiple places under various accounts, the fact remains that your data is still “out there” instead of being behind the firewall in the private database of a company. It’s a peace of mind issue and in some cases is also just a case of who and how much you can trust.
3. Regulation: Then, there is the question of government and taxes. For the near future, the reality is that the govt decides how commerce is done so compliance to tax and other policies will need to be followed. The idea of being independent from the government is not really feasible because of the taxation systems that are very much needed to fund the governments. Web3 may ultimately help the government run more transparent and impartial operations, but adoption of that will be slow.
4. Uncertainty: Finally, as the space shakes out, there is also a risk of technological developments in web3 that will run afoul of the evolving government regulation. In June 2023, the US regulatory body deemed that major cryptocurrency tokens such as Solana etc. are securities and must be regulated. That is not how these protocols have operated till now so it’s a big change for them and their ecosystem. So, until we have a clear idea of how things are going to play out, adoption by the average person may be limited.

If you’ve come this far, then you’re ready to deal with AI & Web3.

Is AI Different in Web3?

The use of AI in analytics and decision-making processes is not “different” in Web3. Traditional companies and organizations employ AI for similar purposes.

Some of these are:

1. Analyzing data for better decision making
2. Automating business processes
3. Generating content on the fly
4. Generating visuals and images dynamically etc.
5. Updating business rules dynamically
6. and so on

The fundamental principles and techniques of AI, such as data analysis, predictive modeling, sentiment analysis, and optimization, are applicable across different domains and industries. They are not limited to either web3 or traditional technology landscapes.

However, the distinction lies not in the AI techniques themselves but rather in how they are employed within the unique ecosystem of Web3 platforms to enhance transparency and community participation.

Let’s identify some use cases unique to Web3.

Key Use Cases for Web3 & AI

I’ll categorize the uses can by the categories we logically think about. Companies such as Stability.ai are making tremendous progress every day in these areas.

Decentralized Decision-Making for DAOs

DAOs, or Decentralized Autonomous Organizations in web3, are organizations that operate based on smart contracts and decentralized governance mechanisms. They allow token holders to collectively exercise their voting rights and make decisions to manage resources without the need for a centralized intermediary.

AI will provide data-driven insights and predictions to facilitate more informed decision-making.

For example, it can simulate the impact of proposed changes using historical data and market trends, This information enables token holders to evaluate the consequences of their decisions and make choices that align better with the DAO’s goals.

AI can also assist with governance optimization by analyzing voting patterns and identifying potential biases or strategic voting behaviors.

So while AI is not doing anything different, we can see how it’s application is unique in a automated web3 environment with DAOs versus in a traditional environment. It is also enabled with better data availability and quality.

Generative AI in Gaming

AI can generate immersive and interactive audio-visual experiences that respond to the game environment in real-time.

As gaming moves towards web3 to facilitate a more ubiquitous foundation layer for user data, generative AI is creating diverse game worlds, levels, landscapes, and environments offering players endless possibilities and exploration in games.

It can learn from existing character designs and behaviors to generate new ones, enabling game developers to populate their virtual worlds with a rich variety of engaging characters. AI can also generate rules, mechanics, and gameplay variations, providing fresh and unique gaming experiences. Among them could be generation of dynamic and adaptive music, and sound effects in games.

All of this will enhance the personalization of a player’s experience. And combined with web3, these outcomes promise to be very powerful, and in control of the users.

Metaverse Commerce & Interactions

Much has been said about engaging in the metaverse. While it’s taken a little bit of a hit in the past year, sensible use cases are coming to the fore everyday with web3 being projected as the foundation.

A few of these use cases address the basics of convenience and a tangible business case. For example, allowing more life like virtual interactions with machines is a boon for both marketing and customer service. The concept of digital twins is also now being extended to virtual worlds where people also interact with each other. This leads to exciting possibilities for the enterprise in the areas of field personnel and customer training that in turn meet crucial compliance and safety needs.

There are marketing use cases as well. Until now, we have seen machines in tradeshows and interacted with them in a limited way using video and other 3D models. With AI now, full interaction and test drives are possible.

Customer service also promises to be enhanced. Imagine that a customer service rep and a customer were able to join the same room, where the customer shows what’s happening with the machine, and the rep can see the same changes on their end. Actions taken by the rep can reflect on the machine as well.

It’s evident that generative AI is going to do much of the heavy lifting that goes into making such a metaverse work by generating needed 3D models, landscapes, and their many variations. It will be supported not just by web3, but also by IoT and other digital capabilities. While these innovations can likely do fine without web3, they could leverage the underlying collaborative layer to drive much higher tech and operational efficiencies.

Better Data Quality Drives Better AI

When we consider the context of Web3, there are some aspects that make the use of AI unique and particularly more powerful in a web3 environment.

Data quality can be a huge advantage in Web3 platforms compared to traditional systems.

For example, we are well aware of the naturally occurring data silos and the resulting data quality issues that exist within each enterprise. These data silos lead to significant and multi-million dollar programs to aggregate and make the data ready for any kind of useful business intelligence and analytics.

Not so in a web3 enabled architecture.

Here’s why.

Single Source of Truth

Web3 enabled applications by design create a single source of truth for the business information.  This eliminates the need to constantly reconcile data across multiple systems.

Contrast that to a traditional enterprise applications landscape where each application must either access a centralized database, or otherwise have their data extracted and staged on a separate data lake system for further analytics and AI. This copying and consolidation of data is the cause of the current woes in the enterprise, sometimes causing up to 60% of the budget being spent on data staging and cleaning.

This unique property of web3 based systems ensures that the data being analyzed by AI algorithms is trusted and current. It minimizes the occurrence of discrepancies or inconsistencies that may arise in traditional systems.

Web3 platforms also automatically lend themselves to fully transparent and comprehensive data lineage and audit. It elevates and amplifies the impact of AI, rather than fixing something that should not have occurred in the first place.

Finally, web3 based applications landscape, by its very nature results in real-time access to data. There are virtually no latencies even as there is a web of intricate cross application business processes are being executed. There are no silos to bridge and transfer data across.

The implications are tremendous for AI models that provide insights for real time interactions such as those with consumers on the web, security breaches, or supply chain responsiveness.

Next Steps

The integration of AI in a Web3 environment is full of possibilities and also has unique scaling benefits by creating a truly digital conglomerate – multiple organizations working efficiently with each other to meet an end to end customer experience.

Just like in the traditional world, AI can enhance decision-making, optimize processes, and provide personalized experiences for users. The inherently better data quality and low latencies on web3 is a significant advantage to achieve AI impact efficiently.

As Web3 continues to evolve, AI will definitely pave the way for a fascinating and secure future.