Knowhere

Vana is the data layer for AI

AI models are ultimately only as good as their training data And AI researchers have run into a problem: we are out of training data. We’ve used up all of the publicly available data (15 trillion words, training dataset size of llama3). To make AI smarter and more capable, we need more data. This is what AI researchers call the “data wall”. With Vana, we’re breaking the data wall by having individuals export and contribute their data to AI models that they own and earn from. Most people don’t realize that you legally own all your data. When you park your car in a parking lot, the parking lot doesn’t own your car - it’s your car. In the same way, when you put your data in an application, your data (writing, emails, posts, photos, audio recordings) is still legally yours. You can always get it back from the platform, thanks to data regulation. Data DAOs on Vana leverage this to get data out of walled gardens. Historically companies have held a monopoly on user data: only Reddit can sell your Reddit data, only Twitter can sell your Twitter data. These platforms each earn $50M-$200M selling user data just for AI training. But with data DAOs, users now have that same power. And they have access to even more data on themselves than any single platform. Some technologists say data ownership slows technological progress. But AI researchers themselves are facing data walls. They need data. Vana’s user-owned approach responds to this need, not only providing data but giving users upside in the AI models their data creates.