USV Lianchuang: Mint means ownership, which is a native business model for both Web3 and AI
You not only tell creators that you like their work, but you also give them a small amount of money and own a copy of the work.Original Author: AVC
Original Compilation: Deep Tide TechFlow
In this article, AVC, co-founder of USV, will explore how Web3 and artificial intelligence are redefining data ownership and business models. Today, personal data is used by large tech companies, and Web3 is changing that landscape. This article will introduce minting as the native business model of Web3 and its potential impact on the AI field. Let’s explore the potential of this new trend together.
We have all been pushed ads, and looking at them, you might think, "How do they know I'm looking for that product?" The answer is that large tech companies utilize AI/ML models trained on our personal data that are incredibly accurate and powerful.
Here are two questions:
The first is that the "data" they use to train the models actually belongs to us, but for the past two decades, we have been handing it over to large tech companies.
The second is that although these models are trained on our data, they belong to large tech companies.
It doesn’t have to be this way, and I believe this situation won’t last long.
Web3 will provide help, and let me explain this possibility.
If you visit zora.co, you will find a social platform that feels like Tumblr, Instagram, Facebook, etc., where you can browse and choose to "like" what you see. But there’s a difference; "liking" on Zora is called "minting." You’re not just telling the creator you like their work; you’re also giving them a small amount of money and owning a copy of that work.
The difference here is that you own one of the pieces you like and have paid a little for it. If creators have thousands of people doing what you did (minting), which is not uncommon on Zora, they can earn a nice income from their work.
And collectors are building their own owned datasets. The datasets are stored on the blockchain and belong to them.
The next obvious step is that companies like Zora will provide collectors the ability to train models on their collections. This turns their collections into training datasets. But the difference is that it’s the collectors who own the training datasets, not Zora.
Soon we will have open-source AI/ML models that can run on our phones. These will be our models, and we can train them on our datasets.
Here is a screenshot of my Ethereum wallet connected to this blog. You can see some collection transactions from the past week or two.
USV Co-founder: Minting is ownership, a native business model for both Web3 and AI
So what’s happening here?
Writers are compensated for their work.
Readers are building their owned datasets on-chain, rather than on Facebook.
The next step is that we have our own open-source models, and we train them on the collections we are building.
These open-source models will help us write, find new reading content, and more. They can inspire us to start a new company, invest in a new company, listen to a new song, and do many other things we want to do.
Returning to Chris Dixon's (A16Z General Partner) words from yesterday's post: In the long run, we still need to establish economic contracts between AI systems and content providers. AI always needs new data to keep up with the times. The world is constantly evolving: tastes are changing, new genres are emerging, and new things are being invented. There will be new topics that need to be described and expressed. Those providing content for AI systems need to be compensated.
There is a path forward for writers, readers, collectors, creators, and everyone.
It starts with us owning our works and allowing others to pay to collect our works.
What makes me optimistic about this is that it’s not just talk and dreams. It’s happening right here on this blog. The tools we need to change how the world operates are already here; we just need to start using them.
