“Big Tech dominates today’s AI economy with its centralized infrastructure,”said Jacob Cantele, Co-founder & CEO of CoinFello. Cantele was a key product leader at Consensys, where he oversaw MetaMask, the leading Ethereum wallet. CoinFello is an AI agent platform launched in March.
Blockchain based AI projects exist, of course, but that corner of the Web3 market is much smaller than the overall blockchain/cryptocurrency market.
The market’s largest AI-linked tokens mostly use blockchains for payments and identity. The strongest current examples are Cortex Labs for direct on-chain inference, and AI infrastructure plays ORA and Ritual. These are available to more sophisticated crypto traders on decentralized exchanges rather than on the big centralized ones like Kraken.
“As of this moment, there hasn’t been major traction for any Web3 uses of AI outside of DeFi or development-stage projects,” said Dr. Ben Goertzel, CEO and Chief Scientist at SingularityNET & CEO at the Artificial Superintelligence Alliance (ASI).
“A key factor is a number of recent research results in the AI space that will allow advanced AI systems to be trained more effectively on decentralized networks, without the need for huge hyperscaler server farms. As these alternate methods of training large neural models and other AI systems become a little more mature, they will enable decentralized networks to rival or exceed what the Big Tech hyperscalers are able to do,” he predicted.
AI-linked tokens and tokenization-related projects have been relatively resilient, with AI becoming one of crypto’s key investment themes.
In the first quarter, AI-linked tokens outperformed as attention on artificial intelligence accelerated more broadly among investors and the need for on-chain financial infrastructure for agents is becoming clearer.
“We are some of the people that are building an alternative to Big Tech AI,” said CoinFello CEO Cantele. “We want users, rather than corporate monopolies, to own and control their own agents. We’re creating verifiable, user-controlled AI agents that operate autonomously in the decentralized economy.”
The nearer a startup gets to true on-chain AI, the harder the engineering and the smaller the market cap. Bigger tokens like Bittensor benefit from incentives and settlement rather than pureplay AI execution.
Some analysts are forecasting that Bittensor could reach $1,338.94 by 2030.
For now, it is only a small, but technically serious subset of Web3 projects that are trying to make AI execution, model ownership, and machine-generated outputs verifiable, tokenizable, and settleable on-chain.
“I have said for a long time that Web3 AI will not succeed just by virtue of being Web3 because there are not enough customers outside DeFi to care enough about data sovereignty, decentralization or the other virtues of the blockchain world,” said Goertzel. “Web3 AI will succeed by being smarter than Web2 AI. I think we are getting closer to that point now.”
Cover art created by the author using Canva.
Benzinga Disclaimer: This article is from an unpaid external contributor. It does not represent Benzinga’s reporting and has not been edited for content or accuracy.