The future envisioned by AI and Web3.0: What synergies will they create?

The future ahead, when we have AI with intelligence beyond that of humans

The term Web 3.0 has many different nuances and cannot be defined in general terms, but if we were to give a common answer that everyone can agree on at present, it would be "decentralized Internet".

It is still fresh in our memory that blockchain was developed from the idea of ​​exchanging information without a central administrator, and the current boom in Web 3.0 began with its application. Smart contracts were devised to allow information transactions on blockchains to proceed smoothly without an administrator.

With the development of this system, which processes fully automatically when certain conditions are met, the idea of ​​Web 3.0 without a central administrator has become more realistic. However, since smart contracts are programs, even obviously fraudulent transfers are automatically processed as long as the conditions are met.

As you can see from the Ethereum split*1, the holes in the smart contract system have led to distrust of the blockchain itself. Therefore, experiments are being conducted to equip smart contracts with AI and have them make the final decision on whether something is right or wrong.

In order to monitor fraudulent transactions, many cryptocurrency exchanges go to the trouble of auditing smart contracts, which is a bit of a misstep. However, in the Web 3.0 culture that dislikes central administrators, manual audits are inevitably incompatible.

In response to such voices from crypto enthusiasts, for example, Bunzz, a smart contract development infrastructure, has begun offering an AI-based auditing service called "Bunzz Audit."

It is said that there are still many challenges to be overcome with AI audits, but considering the rapid progress of AI and the Web 3.0 field in recent years, it is not likely to be a long time before AI audits become commonplace.

Since remittance and settlement processing is operated under strict rules, there is essentially no need for humans to make decisions. Therefore, it would be possible to automate it with AI, as in the case of existing financial systems.

At the same time, if an AI capable of auditing blockchain transactions and dealing with and judging problems is born, another question is expected to emerge: "Are humans really necessary to judge all projects on the chain?"

Currently, various DAOs*2 are being launched on the blockchain, and lively discussions are being held to realize the projects.

However, in order to obtain voting rights, governance tokens*3 are required, and no matter how lively the discussions in the DAO are, there is always a possibility that the opinion of a person who holds a large number of governance tokens will overturn the decision. Can this really be called a fair vote?

Instead, after the project is launched, the generating AI collects data. Humans only help collect and input that data. Finally, the generation AI will analyze the collected data, select the method with the highest probability of project success, and make the decision.

By doing this, it may be possible for the AI ​​to manage and execute the project without any human intervention. Of course, current generation AI cannot do this, but what do you think about fairness?

Examples of AI x Blockchain

AI × Yield Farming

Yield farming is the act of depositing crypto assets on a platform and earning interest in return for providing liquidity to the blockchain. It has become more common among crypto users. By using AI in this process, you can make investments that provide the greatest return on investment while checking the interest rate.

AI x Design Intellectual Property Management

Since generative AI continues to learn based on miscellaneous information, sometimes the generated products infringe on intellectual property rights, which can be a problem. Therefore, a service has been launched that guarantees the originality and value of designs by centrally managing intellectual property data with generative AI and converting it into NFTs with smart contracts.

AI×Play to Earn

Play to Earn allows you to earn crypto assets by playing games. Taking this a step further, a game has been created that turns items acquired during gameplay into NFTs. These items have their own unique story attached to them by AI and are hidden on the blockchain. Of course, since they are NFTs, they can also be bought and sold.

AI x Security

Generative AI has the problem of not knowing what information it has learned and where it came from. Therefore, verification has begun to record the learning path on the blockchain, monitor AI training, and make it possible to trace the decision-making process. By confirming the learning process as transparent and tamper-proof, it should be possible to develop more secure AI.

*1: In 2016, Ethereum was hacked and approximately 6.5 billion yen worth of tokens were stolen. In order to freeze the stolen money, the Ethereum developers intervened in the system and performed a rollback. Some users opposed this and branched off into Ether Classic, causing it to split.

*2: An organizational form that does not have a specific owner or administrator and can promote businesses and projects. Decisions are made by voting via the blockchain, and the system automatically executes them after the decision is made.

*3: A token that comes with voting rights to participate in DAO decision-making.

Related articles

Beyond generative AI: "AGI" and "ASI"

The advent of the AI ​​era: what is the vision for the future of society?