In February 2025, Reddit concluded an agreement of $ 60 million with Google to allow the search giant to use data on the platform to train its artificial intelligence models. Reddit users were notably absent from the discussions, the data of which were sold.
The agreement reflected the reality of modern Internet: large technological companies have practically all our data online and decide what to do with this data. Unsurprisingly, many platforms monetize their data, and the fastest way to accomplish that today is to sell it to AI companies, which are themselves massive technological companies using data to form increasingly powerful models.
The Vana decentralized platform, which began as a MIT class project, is on a mission to make power to users. The company has created a network entirely belonging to users that allows individuals to download their data and govern the way they are used. AI developers can present to users new model ideas, and if users agree to contribute their data for training, they obtain proportional property in models.
The idea is to give everyone an interest in AI systems which will increasingly shape our company while unlocking new data pools to advance technology.
“This data is necessary to create better AI systems,” said Vana's co-founder Anna Kazlauskas '19. “We have created a decentralized system to obtain better data – which is in large technological companies today – while allowing users to keep the ultimate property.”
From economy to blockchain
Many high school students have photos of pop stars or athletes on the walls of their room. Kazlauskas had a photo of the former secretary in the United States of the Treasury Janet Yellen.
Kazlauskas came to put it on that she would have become an economist, but she ended up being one of the five students to join the Bitcoin Club Mit in 2015, and this experience led her to the world of blockchains and cryptocurrencies.
From her dormitory to MacGregor House, she started to exploit the Ethereum cryptocurrency. She sometimes even traveled the dumpsters on the campus in search of thrown computer chips.
“It interested me all around IT and networking,” explains Kazlauskas. “This involved, from the point of view of the blockchain, the distributed systems and how they can move economic power to individuals, as well as artificial intelligence and econometrics.”
Kazlauskas met Abal Art, which then frequented Harvard University, in the former emerging media class, and the couple decided to work on new ways of obtaining data to form AI systems.
“Our question was: how could you have a large number of people contributing to these AI systems using more distributed network?” Kazlauskas remembers.
Kazlauskas and Abal were trying to process the status quo, where most models are formed by scraping public data on the Internet. Large technological companies often buy large sets of data from other companies.
The founders ‘approach has evolved over the years and has been informed by Kazlauskas' experience to work with the company Blockchain Celo after obtaining the diploma. But Kazlauskas attributes his time to help him think about these problems, and the instructor of Emergent Ventures, Rameh Raskar, still helps Vana to think about research questions on AI today.
“It was great to have an open opportunity to simply build, hack and explore,” says Kazlauskas. “I think this ethics is really important. It's just about building things, seeing what works and continuing to iterate. ”
Today, Vana takes advantage of an little -known law that allows users of most large technological platforms to export their data directly. Users can download this information from quantified digital wallets in Vana and pay it to form models as they see fit.
AI engineers can suggest ideas for new open source models, and people can pool their data to help form the model. In the blockchain world, data pools are called Data Daos, which represents a decentralized autonomous organization. Data can also be used to create personalized AI models and agents.
In Vana, data is used in a way that preserves user confidentiality because the system does not expose identifiable information. Once the model is created, users keep the property so that each time it is used, they are rewarded in proportion to the amount of data that their data has formed.
“From a developer's point of view, you can now create these hyper personalized health applications that take into account exactly what you ate, how you slept, how you do the exercise,” explains Kazlauskas. “These applications are not possible today because of these closed gardens of large technological companies.”
IA of Crowdsourced, belonging to the user
Last year, an automatic learning engineer offered using VANA user data to form an AI model that could generate Reddit publications. More than 140,000 Vana users contributed their Reddit data, which contained publications, comments, messages, etc. Users have decided on the terms in which the model could be used and they maintained the property of the model after its creation.
Vana has activated similar initiatives with data controlled by the user of the social media platform X; Sleep data from sources like Oura Rings; And more. There are also collaborations that combine data pools to create wider AI applications.
“Let's say that users have Spotify data, Reddit data and fashion data”, “ Kazlauskas explains. “Usually Spotify will not collaborate with these types of businesses, and there is in fact regulation against this. But users can do so if they grant access, so these multiplatform data sets can be used to create really powerful models.”
Vana has over a million users and more than 20 DAOs live. More than 300 additional data pools have been offered by users of the Vana system, and Kazlauskas says that many will be in production this year.
“I think there are a lot of promises in generalized AI models, personalized medicine and new consumption applications, because it is difficult to combine all this data or to have access to the first place,” explains Kazlauskas.
Data pools allow user groups to accomplish something that even the most powerful technological companies are struggling today.
“Today, large technological companies have built these data of data, so the best data sets are not available for anyone,” explains Kazlauskas. “It is a problem of collective action, where my data is not as precious, but a pool of data with tens of thousands or millions of people is really precious. Vana allows these pools to be built. It is a win-win: users can benefit from the rise of AI because they have the models. Then, you will not end up in scenarios where you do not take advantage of it.
