How AI Can Help Build Web3

Key Takeaways:

• Generative AI, as the name suggests, is a type of AI used to generate artificial content like text, images, audio, and video content. 
• AI applications in Web3 include the deployment of digital collectibles in gaming, NFTs, asset creation, and software development.
• Outside of content generation, AI can also help in driving the Web3 space by streamlining development processes and improving user experience in decentralised apps (dapps).
• While there are still challenges like copyright, accuracy, and creativity, the era of AI has arrived — AI models are transforming businesses and industries alike. 

An Intro to AI-Generated Content

Artificial Intelligence-Generated Content (AIGC) has become extremely popular recently, with applications like DALL-E and ChatGPT producing impressive visual assets and carrying out human-like dialogue, respectively.

Broadly speaking, generative AI is a type of AI used to generate content — such as text, images, audio, and video — by computer models. AIGC is widely regarded as the next stage of content generation, after Professionally Generated Content (PGC) and User-Generated Content (UGC).

PGC is typically produced by creative professionals like graphic designers and animators for brands to use or publish, while UGC is created by end users and directly shared on social media sites like YouTube, Facebook, or Twitter. 

As AI has developed rapidly in recent years, it can now generate various types of content. Some relevant branches of AI are Natural Language Processing (NLP), which researches how computers process and analyse text, and Generative Adversarial Networks (GANs), which aim to generate new data (e.g., images and videos) with similar characteristics to a training dataset.

AI-generated content can help speed up the creative process, and businesses are starting to take notice of its potential to transform the way content is created and how creative teams function across industries. 

Here are potential scenarios and use cases that connect AI and Web3. 

Applications of AIGC in Web3

Text AI and Its Impact on Web3

Text AI refers to the use of AI for generating text. It is a form of NLP that generates human-like text from a given input to be used in a variety of applications like summarisation, dialogue systems, and machine translation. Text generators of today are used to produce original, creative content for various purposes, and there are areas within Web3 where text generation could be highly useful.

With the help of text AI tools, online search can be reimagined and provide a more intuitive way of navigating the Web. ChatGPT’s latest integration with Microsoft’s online search engine Bing has now introduced a chat interface as a way of searching the Web. 

Meanwhile, Google released its own version of the NLP model called Bard, a LaMDA-powered experimental conversation AI text service that helps simplify complex topics and synthesises insights for queries. 

Generative AI Could Change the Way People Search the Web

Generative AI has the potential to change how people filter information on the Web and could potentially reduce dependence on search engines’ advertising models — something many current Web2 users have been wanting to circumvent for a long time. 

Text generation tools allow users to cut through the noise of SEO-generated content when making a query (albeit with human intervention and fine-tuning involved). If search preferences change in favour of text AI tools, search engines could be replaced, meaning less search-related advertising clutter to dig through — a core criteria of Web3 to put the power of technology back into the users’ hands.

In blockchain gaming, there are many ways that text AI can supercharge the creativity and productivity of game developers and artists. By utilising text AI, fundamental video game elements — like dialogue, story, and character composition, amongst others — can quickly be produced and refined, streamlining the creative process by generating ideas faster.

AI NFTs

AI can also help generate images and videos — types of content that can then be minted into NFTs. These AI-generated NFTs are known as Generative Art NFTs, where the artist will first input a set of rules (like a range of colours and patterns), as well as parameters like the number of iterations and degree of randomness. The computer will then generate the artwork within this specified framework.

One example of this is “CryptoPunks” generator Larva Labs, which created the “Autoglyphs” NFT collection. Below are other examples of NFT collections generated with the assistance of AI. 

Generative Art NFTs

Collection	Description	Max Supply	Sample
Autoglyphs	Released by CryptoPunks creator Larva Labs Created by code running on the Ethereum blockchain	512
Fidenza	Created by visual artist Tyler Hobbs Utilises a versatile algorithm that produces a wide variety of curves and blocks	999
Ringers	Created by artist Dmitri Cherniak The artworks are generated by JavaScript and depict various ways of wrapping a string around a set of pegs	1,000
Chromie Squiggle	Created by Erick ‘Snowfro’ Calderon Consists of randomly generated squiggles in nine different style schemes	10,000
Lost Poets	Created by digital artist Pak It is both an NFT collection and a strategy game	65,536

Collection	Autoglyphs
Description	Released by CryptoPunks creator Larva Labs Created by code running on the Ethereum blockchain
Max Supply	512
Sample
Collection	Fidenza
Description	Created by visual artist Tyler Hobbs Utilises a versatile algorithm that produces a wide variety of curves and blocks
Max Supply	999
Sample
Collection	Ringers
Description	Created by artist Dmitri Cherniak The artworks are generated by JavaScript and depict various ways of wrapping a string around a set of pegs
Max Supply	1,000
Sample
Collection	Chromie Squiggle
Description	Created by Erick ‘Snowfro’ Calderon Consists of randomly generated squiggles in nine different style schemes
Max Supply	10,000
Sample
Collection	Lost Poets
Description	Created by digital artist Pak It is both an NFT collection and a strategy game
Max Supply	65,536
Sample

As of 18 January 2023 Sources: OpenSea, CoinDesk, NFT Evening, Crypto.com Research

AI Avatars and Items in Blockchain Gaming

Generative AI models can assist with the creation of game assets at scale in a Web3 environment — anything from avatars, equipment, vehicles, and artefacts. The gaming industry can apply text-to-image generative AI models capable of producing creative assets and content from text descriptions. Within certain parameters, modern language models can also be used to build context around the created assets, such as item power statistics, character attributes, or lore.

AI-generated images and videos are now so advanced that they can even be used to create special effects in blockchain games and virtual products in the Metaverse. For example, Mirror World is a GameFi project that utilises AI-driven virtual “Mirrors” that act as assets for characters in the game. The Mirror assets are fully interoperable in each game, ensuring asset holders will be able to use them in new challenges as they go live. 

Alethea AI’s CharacterGPT project is another example of generative AI at play. It features a multimodal AI system called CharacterGPT to generate interactive AI characters from a text description, therefore enabling text-to-character creation. The interactive characters can have distinct appearances, voices, personalities, and identities based on different natural language descriptions.

The characters can be tokenised on the blockchain, and their owners can also customise their personalities and train their intelligence, as well as trade and use them across various other dapps on Alethea’s AI Protocol. The proposed use cases of these interactive characters include Digital Twins (virtual models designed to reflect a physical object), Digital Guides, Digital Companions, Virtual Assistants, as well as AI Non-Player Characters (NPCs).

AI Can Help Find Bugs

AI can help in streamlining the development process when it comes to building Web3 infrastructure and applications. 

For instance, AI applications are used for debugging code. Using AI, ChatGPT has shown, to some extent, the ability to not only read and write code but find bugs in the code.

Some crypto professionals have now started using the AI-powered programme for simple code audit tasks: Developers at smart contract auditing firm Certik used ChatGPT for “quickly understanding and summarising the semantics of complex code snippets.” 

Take your research to the next level and read our in-depth analysis of AI’s applications in the Web3 space in our latest feature report.

Final Words: Challenges, Risks, and Outlook for AI Use in Web3

With AI, the possibilities are endless and only limited by the users’ imagination. Even in their early phases, AI models continue to demonstrate their capabilities in transforming businesses and even industries. With low barriers to entry resulting in widespread adoption, it is likely that AI will become our future way of life in this digital world. However, there are also some challenges and risks that come with this type of technology.

One challenge could be pushback against AI-generated content by consumers and organisations. For example, Getty Images, a major stock photo website and platform, has disallowed the upload and sale of illustrations generated using AI art tools. Copyright concerns are being cited as the reason, as some of the AI-generated images reproduced copyrighted content, with the original artist’s watermarks still visible. 

Another challenge for AIGC is quality concerns. Stanford professor Andrew Ng produced an example where ChatGPT erroneously explained how an abacus is faster than a GPU, which is thankfully not true. 

Hitting very close to home for most in the space is evidence that this technology is starting to disrupt the workforce. However, it is a misconception that AI will replace humans at work. In fact, it can actually create new opportunities in existing and emerging markets: It is likely that AI will either help augment jobs, or new types of AI-related jobs will be created, with some upskilling required.  

A quote famously attributed to author William Gibson might have described the future of AI best: “The future is already here — it’s just not evenly distributed.” That can also be said of the intersection between AI and Web3 today.

Read the full report on AI and its use cases, including more examples and references, here.

Tracking Web3 and AI Tokens in the Crypto.com App

The rise of AI technology has led to the emergence of new AI tokens that are quickly gaining popularity. To cater to this interest, the Crypto.com App has added a new category on Track Coins so users can easily follow the top AI tokens. Check out all of the categories here. 

Due Diligence and Do Your Own Research

All examples listed in this article are for informational purposes only. You should not construe any such information or other material as legal, tax, investment, financial, or other advice. Nothing contained herein shall constitute a solicitation, recommendation, endorsement, or offer by Crypto.com to invest, buy, or sell any coins, tokens, or other crypto assets. Returns on the buying and selling of crypto assets may be subject to tax, including capital gains tax, in your jurisdiction. 

Past performance is not a guarantee or predictor of future performance. The value of crypto assets can increase or decrease, and you could lose all or a substantial amount of your purchase price. When assessing a crypto asset, it’s essential for you to do your research and due diligence to make the best possible judgement, as any purchases shall be your sole responsibility.