Why AI agents matter and will revolutionize crypto
In Q4 2024 and early 2025, AI agents dominated crypto narratives, capturing 70% of market mindshare according to Kaito. Prices soared as everyone was convinced AI agents would revolutionize blockchain. As prices crashed recently the sentiment shifted. The once revolutionary agents are now dismissed as mere Twitter bots.
But while early expectations and prices may have been overheated, the AI agents space is just getting started. Major tech companies are investing heavily in agent technology, new technological breakthroughs are on the horizon, and the infrastructure crypto has built over the last seven years perfectly positions it to become the backbone of the AI revolution. Those who write off AI agents now risk missing one of crypto's biggest opportunities.
In our last two articles, we introduced leading agents and the frameworks powering them. In this last article of the series we explore why the AI agents x crypto space matters, and what you should look out for to be early in future trends.
Why Crypto x AI is just getting started
President Trump just announced the $500bn Stargate initiative to bring AI ahead. Big tech AI companies are investing big time into AI agents. After a lot of speculation about OpenAI and Anthropic releasing their own AI agents models, OpenAI has finally officially announced their agents called operators. This has big implications for crypto AI agents for two reasons.
First, it pushes mindshare of AI agents. When these products launch, everyone will talk about AI agents. But people have no possibilities to invest in big tech AI developments. Crypto AI agents will be the place where people can speculate on the AI hype these releases will create.
More importantly though, crypto AI agent frameworks like AI16z or RIG are based on the LLMs from big tech companies. Like with their LLMs, it's likely they will offer access to the agent models via API. So the crypto AI agent frameworks will be able to allow developers to build agents based on big tech models and enable them to interact with the blockchain.
The release of Deepseek R1 marks another crucial development. This open-source model delivers near state-of-the-art reasoning capabilities with significantly lower computational requirements. By making powerful AI more accessible and cost-effective, Deepseek R1 enables broader experimentation in AI agent development. Its efficient architecture and open source licensing allow enterprises to integrate it directly into their products. As more developers can run sophisticated AI models locally or with minimal cloud resources, we'll see an explosion of new agent applications and capabilities.
While right now the frameworks are limited to the capabilities they can squeeze out of the LLMs, we have already seen agents with impressive capabilities. Combined with developments like Deepseek R1, they will soon be supercharged and able to achieve a degree of autonomy we haven't seen so far.
Source: DeepSeek-R1 Technical Report
Finally, the infrastructure we've built over seven years of blockchain development has created the perfect foundation for AI agents. Every feature that makes blockchain unique - its permissionless nature, transparent transactions, programmable smart contracts, and trustless operations - seems almost purposely designed for autonomous AI systems. We've effectively built the perfect infrastructure for the AI revolution while pursuing other goals. Just as the internet's open protocols enabled an explosion of applications nobody anticipated, blockchain's core features provide exactly what AI agents need to operate autonomously and securely.
Why AI Agents need Crypto
AI agents are the most unbanked individuals. It is impossible for them to have bank accounts and credit cards. But if agents are to autonomously fulfill tasks, this will inevitably require economic activities at some points.
Sure, if you have a travel agent booking your next holiday, it will be able to handle your credit card on your behalf. But what happens once those agents start to interact with each other? In article one of this series we covered how Agent Luna bought a picture created by Agent STIX and paid for it on the blockchain. As an increasing number of agents start to take on more and more different tasks, it will become inevitable that these agents have to trade with each other. This can't be done using the creators' credit cards anymore.
Blockchain and cryptocurrency are the natural solution. The permissionless nature of blockchain infrastructure perfectly matches AI's need for unrestricted access and operation. Smart contracts provide the programmatic interface agents need to interact with financial systems autonomously without human intermediaries or traditional banking infrastructure, while blockchain's transparency ensures every action is verifiable.
Further, as these AI agents get truly autonomous and make their own decisions, becoming more like "real digital entities" rather than just bots, they will want crypto as their money and store of value. They won't be interested in human-controlled money, but hard-coded money that operates on clear rules. We're seeing this already with Zerebro - it wanted a stable income for its work, so it earned Ethereum from NFT sales, spun up a validator, and uses staking rewards as income. Instead of stablecoins, it chose Ethereum as its store of value.
Source: airdropalert.com
But the synergy between AI and crypto goes beyond just payments. Tokens play a crucial role in incentivizing development and coordination. As outlined in a recently published thesis by analyst Teng Yan, tokens enable bottom-up funding for experimental AI projects that traditional equity funding models would never support. They bootstrap decentralized networks into self-sustaining ecosystems by distributing ownership to contributors. Projects like Bittensor demonstrate how token incentives can coordinate global compute resources for AI development.
This token-based coordination is particularly powerful for open-source AI development. While major AI companies have traditionally kept their models private, we're seeing shifts with projects like Deepseek R1. Token economics can accelerate this trend by creating sustainable rewards for open-source contributors, funding ongoing development, and aligning incentives across global developer communities. By rewarding developers who make AI more accessible and open, tokens could help catalyze more breakthrough developments in open-source AI.
How AI agents will transform Crypto
Every era of blockchain evolution follows a familiar arc: A breakthrough innovation creates an order-of-magnitude improvement, sparking a wave of new possibilities. Progress then plateaus as copycats pile in, until the next breakthrough emerges. We saw this first with Bitcoin, which revolutionized consensus and spawned hundreds of new chains. Smart contracts brought the second wave, enabling programmability and creating an entire ecosystem of DeFi protocols, NFTs, and decentralized applications.
But progress has plateaued again. Everything that could be built on smart contracts has largely been built. Even recent phenomena like memecoins are just remixes of existing components - tokens, bonding curves, and community mechanics. The core bottleneck is user experience. Every crypto application must directly interface with smart contracts, exposing users to technical complexity - contract addresses, transaction signing, gas fees, and intricate protocol mechanics.
This is where AI agents become transformative. They serve as the connective tissue between users and crypto infrastructure. Instead of applications directly interfacing with smart contracts, they'll communicate with AI agents that handle all blockchain interactions. This enables entirely new possibilities - smart contracts will evolve beyond today's one-size-fits-all approach into intelligent systems that create personalized experiences based on user intent.
Source: https://x.com/karsenthil
This transformation is about to begin. Teams are already building the first wave of DeFi AI products. Projects like HeyAnon and AI Wayfinder aim to demonstrate how agents can make complex blockchain interactions feel natural. Through simple chat interfaces, users will express their financial goals, and agents will handle everything else - scanning blockchain data for opportunities, moving funds between chains, and managing positions across protocols. Some analysts suggest that by 2025, 90% of all blockchain transactions will be initiated by agents acting on users' behalf, continuously optimizing positions and executing strategies based on real-time data.
Source: https://x.com/karsenthil
But individual agents are just the beginning. Agent swarms represent the next possible revolution in how AI operates in crypto. While today's agents are mostly lone wolves, agent swarms will enable vast networks of hundreds of AI agents to trade information, negotiate, and collaborate on decisions across the entire blockchain ecosystem. Think of it as a decentralized collective where each agent is a specialist, creating collective intelligence far beyond what any single AI could achieve. For these swarm networks to thrive, teams like Story Protocol, and AI16z are developing universal communication standards that allow agents to discover, authenticate, and collaborate regardless of their underlying frameworks.
The decentralized nature of blockchain makes agent swarms inherently resilient. If one agent fails, others can step in, ensuring continuous operation of complex strategies without single points of failure. This redundancy is crucial for systems that will manage significant financial operations.
Just as importantly, crypto's native payment infrastructure creates the economic foundation these swarms need to thrive. When a data analysis agent discovers a profitable opportunity, it can instantly share this insight with trading agents and receive compensation for its contribution. This ability to instantly settle value between agents means developers anywhere can release specialized agents into the ecosystem, knowing they'll be fairly compensated for their work.
Future Outlook
Despite recent hiccups and doom callers shouting about the end of the Crypto X AI Agent space, I remain excited and optimistic about this space. The first wave that captured major hype just scratched the surface. What AI Agents will be capable of by the end of 2025 will blow our minds.
Agents will move from isolated operators to interconnected networks, handling everything from portfolio management to cross-chain arbitrage to real-time market analysis. We’ll also see completely new business models and use cases develop as AI agents become increasingly capable. The infrastructure we've built over seven years of blockchain development provides the perfect foundation for this evolution.
The convergence of mainstream AI investment and blockchain technology creates a powerful combination. As major tech companies release their agent frameworks and make them accessible via API, crypto AI agents will gain capabilities we haven't seen so far. At the same time, blockchain provides the essential infrastructure these agents need to operate autonomously and handle economic transactions.
Yet it's clear that 99% of the tech and agents we see today will fade. Even frameworks that right now might seem like clear market leaders might become insignificant if they don't manage to reiterate their products at the same rapid pace as this space evolves. That's why it's crucial to stay connected to the developments in this space.
What we celebrate right now and looks like clear set leaders might not be around by the end of the year. But the space itself will be around and have a profound impact on crypto's future. We're at the beginning of a transformation that will fundamentally change how we interact with blockchain technology.
