A massive economic shift is underway as AI becomes more and more capable of doing valuable economic work. In its current state, AI is created by centralized companies: big tech companies train AI models, fully own them, and sell access to them to other companies. These large tech companies have already trained AI models based on your public work, writing, artwork, photos, and other data, combined with everyone else’s data. They have started earning billions of dollars a year from it, with OpenAI, for example, hitting $3.4B in revenue this year.

Also Read: How No-Code Machine Learning Platforms Are Revolutionizing Data Science for Non-Experts

There is a lot of controversy over AI, largely due to data ownership. If you train an AI model on an actor, should the studio own it, or should the actor? If you train a video model on the top YouTube creators, shouldn’t the YouTube creators be compensated in some way? This question has become front and center as AI researchers have run into the data wall, where the publicly available datasets have already been used up to train leading AI models, but the AI models need even more data.
The system is clearly broken—AI companies need more data, which users have but don’t even realize they legally own. One approach to solving this is decentralized AI, where a large number of people can contribute to creating and owning AI models. One important technology in making this possible is Data DAOs, which allow users to aggregate their data for the purposes of training AI models.

Data DAOs are decentralized entities that allow users to pool and govern their data, rewarding contributors with a dataset-specific token that represents ownership of the particular dataset. The DAOs are completely run by the community members involved within the DAO, with validators helping to verify data within a data liquidity pool (DLP) to confirm that the data contributors are adding value before letting them contribute. The DAO can have full control over the dataset and can choose to rent it out or sell anonymized copies. Reddit data, for example, could even be used to seed new, user-owned platforms, complete with friends, your past posts, and other data, ready-to-go on the new platform.

Also Read: AiThority Interview with Paul Fipps, President, Global Industries and Strategic Growth at ServiceNow

Data DAOs don’t just benefit users—they also advance AI progress, making it possible to build AI like open source software in a way that benefits everyone who contributes. Open source AI is struggling to find a viable business model: it is expensive to pay for GPUs, data, and researchers. And once the model is trained, if it is open source, there is no way to recoup these costs. The technical architecture of data DAOs can be applied to AI model DAOs, where users and developers contribute data, compute, and research in exchange for ownership of the model.

Currently, big tech steals user data to train AI models, who are ultimately profiting from our data. If someone has access to your data, they could create an AI version of you that is economically valuable and could financially gain from your data. Through Data DAOs, the individual has full control over their data through a concept called “non-custodial data” that is similar to a non-custodial wallet—but for a user’s personal data that adds a layer of privacy to keep your data safe.

As AI becomes a greater and greater part of our daily lives, it will become a main source of truth in society. Having a single company control that source of truth presents risks. Google Bard, for example, which originally had imposed censorship and moderation which rewrote history to show more diversity, was taken down and framed as a mistake. But if there wasn’t another source of truth, society may have just taken that as truth.

Bringing data out of walled gardens and enabling community-owned AI is critical to ensuring that AI benefits everyone, and doesn’t run into the data wall that AI researchers have hit today. Data DAOs built on Vana, like the Reddit Data DAO launched in April, and the upcoming Twitter and LinkedIn Data DAOs, offer a path towards a user-owned alternative.

Decentralized AI offers a path towards powerful AI that is community-owned, offering the best of both worlds. While the default option today is centralized AI, where a single company controls a very powerful AI model, decentralized options are rapidly growing and offer a credible, user-owned alternative. The future of AI may well be one where the power of these transformative technologies is placed back into the hands of the many, rather than concentrated in the hands of a few.

Also Read: What is Return on AI – and How Do Companies Measure It

[To share your insights with us as part of editorial or sponsored content, please write to psen@itechseries.com]

AI Inspired Series by AiThority.com: Featuring Sarah Wieskus, Intel’s GM Commercial Client Sales