The rapid adoption of AI agents is driving businesses to not only implement them as full-time assistants but sometimes to redefine their core approach. This is particularly the case for Itheum, which has taken a journey from a data brokerage platform to a multichain data tokenization infrastructure for AI agents with specific industry use cases.
⚡️We’ve seen the rise of AI agents in many areas but what about music?
— Cointelegraph (@Cointelegraph) February 3, 2025
Check out the AMA recap with @itheum to get the details on them, plus on Data NFT and NFMe ID tech unlocking new data economy.https://t.co/GJxaIF1NGq
Co-founder Mark Paul shared the story of the project, the unique technology it is built on, and his views on decentralized AI in the latest Cointelegraph AMA.
Not your regular NFTs
Data NFTs are one of the concepts revolutionized by Itheum. ”Don’t get too confused by the fact that we say NFT,” Mark noted, ”because we just see it as the most effective type of technology to tokenize data. You can actually tokenize anything this way - health, social media data or anything else, but right now we are focusing on the music industry.”
He explained why Itheum decided to target this particular market: ”It’s a real problem area that needs to be solved, and AI is actually going to disrupt this industry. And finally, indie musicians get a lousy deal. They get screwed in every way. So, we can bring all of that together with a blockchain-based platform that can provide value to them and their fans.”
As Mark went on to explain, musicians can tokenize their albums as NFTs with all the usual properties, but thanks to the unique features of Data NFTs, they have more flexibility. ”You can change the data, add a new song, or change the metadata. For example, if you want to give royalties to an artist who inspired your track, you can embed them directly into the token. You can’t do that with regular NFTs,” Mark pointed out.
AI agents as tireless music managers
Mark mentioned Sigma as one of the AI music agents using Itheum technology. ”It’s built by a separate team and designed to work in synergy with real-world musicians to help them with every aspect of their career – be their promoter, community manager, listing manager, royalty handler.
Sigma has a wallet, and when you send a music data NFT into that wallet, it can autonomously unpack it. It can then stream the music, list it on platforms like Tensor, and manage distribution and royalties. It can even organize collaborations by tagging musicians on social media, bringing artists together for remixes, and tokenizing new tracks to start the cycle again. Sigma can do this 24/7 while you’re sleeping,” he elaborated.
According to Mark, Itheum helps launch at least one music album a week and currently has nearly 60,000 music entities in circulation. ”The musicians provide their tracks and we work with the artist to then handle the rest to launch the music simultaneously on multiple channels (directly on Itheum’s socials, Solana’s Drip Haus, Sigma’s music streaming app sigmamusic.fm), splitting the royalties equally,” he noted.
Decentralized AI meets human curation
”It’s obvious that AI is going to take jobs across industries, and it’s going to be better than us,” Mark said. ”But one thing we can do as humans is to offer our services to AI from a kind of workforce perspective.”
That idea led Itheum to create the AI Data Workforce, which allows individuals to curate and refine AI-generated content and AI agents to tokenize creative work. Users can join by minting an avatar-like NFMe ID vault, which is essentially a Data NFT, get initial BiTz XP points, and start working on their onchain reputation by completing tasks from AI.
”Say AI produces two music tracks. You log in with your NFMe ID, prove your reputation, and vote for one of the tracks with your BiTz XP,” Mark explained. ”This builds your reputation while providing a valuable service.”
NFMe IDs also serve as encrypted digital vaults for personal data. ”For example, you could store a zero-knowledge proof of your age,” Mark noted. ”When you vote on AI-generated content, the AI could query that information to verify that the preferences are relevant without revealing personal data. Of course, all interactions are consent-based. If an AI requests data, it can only access it if you engage with the task.”
Mark believes that AI will remain mostly centralized, at least for some time: ”It’s naive to think that decentralized AI will take over. But the advantage of Web3 communities is that they can get people involved and build the community rewards for them. This is in contrast to centralized companies like OpenAI, where I provide data that improves the model and they charge me more for it.”
He then touched on the topic of Itheum’s reward system: ”Right now, rewards are based on liveness reputation. But we plan to introduce a model where rewards are also based on how much valuable data you share through NFME IDs.”
Looking ahead, Mark expects the AI Data Workforce model, and AI agents in general, to benefit multiple industries: ”Music is just one example. It could be the financial industry, personal assistants, trip planners, etc. Our goal is to build technology that allows AI agents to do things more efficiently. And as a result, bring more value to people.”
Disclaimer. Cointelegraph does not endorse any content or product on this page. While we aim at providing you with all important information that we could obtain in this sponsored article, readers should do their own research before taking any actions related to the company and carry full responsibility for their decisions, nor can this article be considered as investment advice.