Key Takeaways
- AI is dominating global tech narratives, and crypto is keen to tap into that energy.
- Projects like TAO, ai16z and Virtuals are riding a wave of renewed interest and price action.
- While the tech still lacks mainstream use cases on-chain, speculation is kicking into high gear.
- AI agents and decentralised infrastructure may ultimately unlock this crossover use case.
The AI + Crypto Crossover
AI is the biggest tech innovation in the world today. It is being adopted by every major company, integrated into countless platforms and reshaping how people work, learn, and communicate.
Crypto, by nature, is always looking for the next big integration – and AI is a logical fit. The crossover feels inevitable.
Here’s a simple way to think about it:
AI can’t open a bank account for itself, but it can create a blockchain wallet and start transacting. That opens up new possibilities, like AI paying people – or even other AIs – in real-time and without permission.
We are still early. But the building blocks are already being laid.
AI Infrastructure Making a Comeback
Q4 of 2024 saw the blockchain–AI crossover explode in hype and price movement. Most of the momentum peaked in early January 2025 as speculation cooled and the broader crypto market, including Bitcoin, endured a pullback.
But as the market has rebounded, led primarily by Bitcoin reaching a new all-time high, certain AI projects are staging impressive comebacks. TAO, ai16z and Virtuals are among the standouts, demonstrating resilience and price movement that demand attention.
Price increases of top AI projects since their mid-April lows
These projects rebounding is not just them riding a second wave of AI hype. Each is contributing something distinct to the AI x blockchain ecosystem.
TAO – Decentralising Machine Learning
Bittensor (TAO) is building a decentralised neural network where machine learning models compete for influence and rewards. The system operates as an open marketplace where contributors upload models that get ranked based on performance. Top models receive TAO tokens, incentivising innovation in a trustless environment.
Bittensor’s biggest strength is the architecture, fully on-chain, transparent and permissionless. TAO does not just decentralise computation. It decentralises the economics of AI training itself.
It is earning a growing reputation as the ‘Ethereum’ of decentralised AI infrastructure, a foundational layer that other projects may eventually build upon.
Chutes, the most popular project on Bittensor, operates as a decentralised alternative to AWS Lambda. While Lambda transformed how developers run backend code without servers – cutting costs for web-based businesses – it’s still centrally controlled by Amazon and not specifically built for AI. This is where Chutes shines – it is designed for running AI workloads in a transparent, community-powered way.
If AI agents like this continue to gain traction, they could disrupt existing models and redefine how we interact with the internet.
ai16z/ElizaOS – Where Meme Meets Model
ElizaOS is part social experiment, part emerging protocol. Its core value lies in token-gated AI interactions, where chatbots and digital agents respond only to verified token holders. The project provides a framework for developers and individuals to deploy or engage with specialised AI agents that maintain intelligence across different platforms.
The project has gone through a rebranding process from ai16z to ElizaOS, but the native token remains as ai16z.
It does not boast major technical breakthroughs, but its blend of meme culture, AI themes and scarcity-driven mechanics has sparked strong speculative interest.
Interestingly, ai16z’s Eliza framework has gained massive attention from developers around the world. In December 2024, it became the most popular project on GitHub, the main website where developers share and build open-source software. This means a lot of people are not just watching the project, but actively exploring how it works, copying the code, and building their own versions. That kind of activity shows there is real interest in what ElizaOS is creating, beyond just hype. Companies are already beginning to leverage Eliza’s architecture to deploy 24/7 help bots that don’t just repeat a script – but become part of a community.
Virtuals – AI Agents That Live On-chain
The Virtuals Protocol project explores one of the more futuristic ideas in the AI and crypto space – autonomous agents that live directly on the blockchain. These are not just smart bots, they are programmable digital entities that can create content, hold wallets, execute transactions and interact with other users or AIs in real time.
Each Virtual is essentially its own on-chain identity, powered by smart contracts and designed to evolve. Some generate music, others trade crypto and a few are being trained to operate full decentralised services.
This concept taps into the growing excitement around AI agents, but with a crypto-native twist, giving them ownership, autonomy and economic incentive. As interest in agent-based systems grows, Virtuals are gaining traction both in developer circles and on-chain communities, with recent price action suggesting the market is paying attention.
A popular example of Virtuals’ potential is the aixbt agent – an AI agent designed to scan X (formerly Twitter), Reddit and other social media sites to identify crypto market trends. Given how crucial sentiment is in crypto, aixbt’s real-time updates – powered by millions of posts – give its insights a significant edge.
And of course, being an AI agent that spends eternity trawling through crypto social media, it knows its way around a meme or two.
Speculation Rather Than Product
While AI has already transformed many industries outside of blockchain, from content creation to logistics, its role within the crypto ecosystem is still being figured out.
So far, most AI and crypto projects are long on vision but short on real-world usage. They are experimenting, attracting attention and building hype, but very few have delivered working products that solve actual problems at scale.
This is not necessarily a bad thing. Crypto has seen this play out before. In the early days of decentralised finance (DeFi), many protocols were clunky or unproven. But from that experimentation came giants like Uniswap and Aave.
But as it stands, what we mostly have is infrastructure and speculation. The real use cases, if they emerge, are still ahead.
Disclaimer: The information on Swyftx Learn is for general educational purposes only and should not be taken as investment advice, personal recommendation, or an offer of, or solicitation to, buy or sell any assets. It has been prepared without regard to any particular investment objectives or financial situation and does not purport to cover any legal or regulatory requirements. Customers are encouraged to do their own independent research and seek professional advice. Swyftx makes no representation and assumes no liability as to the accuracy or completeness of the content. Any references to past performance are not, and should not be taken as a reliable indicator of future results. Make sure you understand the risks involved in trading before committing any capital. Never risk more than you are prepared to lose. Consider our Terms of Use and Risk Disclosure Statement for more details.
