Remember those writing prompts from high school: Describe your favorite cookie.
Your teacher told you to write it as to an alien, a being who had never encountered a cookie before, which meant touching every sense: sight, hearing, smell, touch, taste. You may not have realized it then, but describing something in a way that allows people to get a clear idea is quite difficult.
Let me try to describe Matheus Pagani, founder and CEO of Venture Miner. Matheus is a man with light caramel skin and dark brown hair. Although her hair is very short, you can tell that it is curly. He has a thick dark brown, almost black, beard that connects with a mustache. His eyes are dark brown behind thin wire glasses. His lower lip sticks out a little further than his upper lip, giving him a look of confidence, but not arrogance.
Can you already imagine it? How sure are you?
Oh yeah, and he’s Brazilian.
I understand?
Let’s see what Matheus Pagani is really like.
Is this what you came up with from my description? I doubt it. Every time I told you he was Brazilian, did you put him in brightly colored accessories and a feather headdress? Something like this?
If so, check your bias, but you are also thinking like an AI. That’s what ChatGPT came up with from the “some Brazilians having fun” message. Pagani showed off this and other examples spit out by our generative AI (Italians having fun sitting around long tables with multiple generations eating pizza) during the AI2Web3 Bootcamp in New York in early December.
The bootcamp, led by Pagani and Build City, brought together 59 participants of all skill levels to learn how the two most popular (and often misunderstood) technologies can be combined to create useful products and services. Pagani used a version of the high school assignment to explain how and why AI made the important leaps that have kept us all excited and on edge in recent years. Previously, text data was largely only used to train AI, and as the exercise highlights, that only goes so far. But mix text information with visual data and you get a more complete picture.
And understanding this, getting hands-on with AI and blockchain technology to understand its core components is what the bootcamp was all about. For Pagani, these skills will be relevant to almost everyone (engineers, technology users, journalists, artists, doctors) very soon.
“We want to bring together brilliant minds from all backgrounds to work with AI and Web3, as bringing together their multiple perspectives can uncover new use cases that we would never imagine with just a specialized Web3 or AI mindset,” Pagani said. “Today we have tools that allow any non-technical enthusiast to create practically functional applications and systems simply with “plain English”, so what matters is bringing together people who are passionate and interested in solving problems along with the right education. When you have this combination, you just need to light the match and watch it burn.”
Amazing building
What makes the intersection of these two technologies so exciting is how much can be built in such a short time without any prior technical experience.
Not only will AI get complete codebases with the right message, but the crypto industry is also creating tools to help make development at the intersection of the two more intuitive and accessible.
For example, Coinbase, which sponsored the bootcamp, launched AgentKit in November. The framework allows developers to create AI agents with their own crypto wallets, allowing them to autonomously interact with blockchain networks. This could be used to create a team of agents who can monitor the markets and automatically execute trades based on predefined rules and guardrails.
“One day, AI agents will have their own cars and operate their own taxi service that customers pay in cryptocurrency and then use those cryptocurrencies to purchase repairs,” Lincoln Murr, associate product manager at Coinbase, told attendees.
Coinbase currently has an ongoing grant program to build with AgentKit. “What you build doesn’t have to be useful; we have a tendency towards interesting things,” Murr said at the bootcamp, hoping to inspire projects and applications that no one has thought of yet.
Ora Network also has an interesting model for developers looking to create AI-enabled Web3 applications or vice versa. The network allows developers to use large current language models, including Llama3 and Meta’s Stable Diffusion, but also allows them to build their own models and offer a so-called initial model offering (IMO) to crowdfund their continued development.
“In AI, right now the winner takes all, but with this model, we are enabling crowdfunding of building and training AI, so that people can have a share of the models, which is empowering if we believe these models will work. lead the partnership in a decade,” Alec James, Ora’s partnerships and growth leader, said during the bootcamp. “If that’s the case, we’ll want that development spread out.”
Near, Fleek and Alora were also among the companies that sponsored the bootcamp and presented their various tools and programs to build at the intersection of these two innovative technologies.
Can developers do anything?
During the last day of the bootcamp, nine teams presented working prototypes for projects combining Web3 and AI. These projects ranged from artificial intelligence assistants aimed at helping you choose gifts, deliver orders or diversify your financial portfolio to applications to help crypto traders generate memecoins with high virality potential.
Jackie Joya, a participant who had flown in from San Francisco, said the boot camp really inspired her to keep building. With a background in animal science, Joya is still new to engineering, but she was surprised at how much a novice could build with the tools available.
Other participants, at all levels, said similar things. Choudhury Imtiaz, a market researcher from Bangladesh, who is in the US on an H-1B1 visa waiting for a placement, had not heard of Web3 before the bootcamp, but was able to present a team project on the final day. And Isayah Culbertson, who has worked as an engineer for separate cryptography and AI projects, was able to learn skills to build with both, which she believes has the potential to change the world for the better.
“I see the combination accelerating research and development in so many different fields, while allowing for a more equitable distribution of the wealth generated from that research and development,” he said.
