Will AI Agents Be the Next Crypto Trend? Moving from Narrative to Practical Use Is Key
TechFlow Selected TechFlow Selected
Will AI Agents Be the Next Crypto Trend? Moving from Narrative to Practical Use Is Key
We have the tools, and we have the Builder. What we need most now is—discipline.
Navigating Web3 tides with focused insightsAuthor: Alec Goh, Head of HTX Ventures
Building a Sustainable Crypto + AI Value System
In the digital asset space, I've lived through multiple market cycles—most driven by "narratives" and "technological progress," but few truly game-changing.
Over the past six months, I’ve come to believe that the convergence of AI Agents and Web3 has the potential to be one of those rare transformative shifts. But only if the industry moves beyond early-stage hype and focuses on delivering long-term value for end users.
We’ve seen this story before. In 2017, the ICO boom vastly outpaced product development; in 2021, NFTs exploded into the mainstream while most platforms still lacked foundational infrastructure. Both cycles shared one common trait—the bubble rose fast, and the correction followed just as quickly.
The surge in AI Agent tokens follows the same pattern. Dozens of projects branded as “Crypto + AI” have launched, rapidly attracting massive capital and attention. Some reached billion-dollar valuations within weeks. Yet again, hype alone cannot sustain an ecosystem. Once investor enthusiasm fades, user engagement drops—and already, most early projects are losing long-term relevance.
Yet amid this volatility, something fundamentally different is also taking place.
In Q4 2024, DeepSeek achieved a critical technical breakthrough that caught our attention at HTX Ventures. By using reinforcement learning without relying on human-labeled data, they successfully trained large-scale models. This dramatically reduces the cost of developing autonomous AI agents, making scalable and low-barrier deployment of AI Agents in Web3 a realistic possibility for the first time.
This shift is significant. If crypto is to fulfill its promise as a decentralized, programmable financial infrastructure, intelligent agents with autonomous decision-making capabilities will become an essential foundational layer. However, we are still far from realizing this vision.
My concern is that the industry may once again chase trends without laying solid groundwork. True, promising frameworks are emerging: Eliza is building composable agent tooling across multi-chain ecosystems like Ethereum, Solana, and TON; launchpads such as Virtual and Clanker are accelerating token issuance; application-layer projects like GRIFFAIN and NEUR are exploring asset management and governance tools.
But these remain early signals—no clearly validated system or ecosystem has yet formed.
Too many AI Agent projects remain speculation-driven rather than sustainability-focused. If this continues, we risk repeating the old cycle of capital flight, unmet expectations, and stalled development.
That would be a tremendous loss.
Because the demand for these applications is real. DeFi strategies too complex for most users can be automated via intelligent agents; on-chain governance can leverage agents to analyze proposals, compare historical outcomes, and vote according to user-defined values; cross-chain operations can be executed autonomously based on user intent.
These aren’t hypotheticals—they’re real needs emerging today. What’s missing isn’t imagination, but consistent execution, user retention, and viable token economics.
So where do we go from here?
First, we must evaluate AI Agent projects with the same rigor applied to DeFi protocols.
This means moving beyond vague narratives toward disciplined project logic, asking fundamental questions:
● Who are the actual users?
● Does it deliver indispensable utility?
● Where does revenue come from?
● Can the protocol survive after token incentives dry up?
Protocols must demonstrate not just technical potential, but real-world use cases, stable revenue models, and measurable impact. That’s the crucial difference between “proof-of-concept” and “worth building.”
Second, beyond applications, the industry needs to build the infrastructure required for successful AI Agent deployment—agent frameworks, data layers, compute resources, and more. HTX Ventures will continue closely monitoring developments in this space, as some of these components may become as essential for future systems as EVM compatibility was for L1 blockchains.
Finally, we need to foster an industry culture that rewards iteration and transparency.
No team gets everything right on the first try—but those committed to continuous refinement and improvement are most likely to lead the next phase of Web3.
The convergence of AI and crypto is not a “narrative”—it’s an “evolution.” But to truly seize this opportunity, the industry must shift from “capturing attention” to “creating value.”
This isn't just HTX Ventures’ investment philosophy—it's a conviction forged through years of experience and forward-looking judgment.
We have the tools. We have the builders. What we need most now is—discipline.
Join TechFlow official community to stay tuned
Telegram:https://t.me/TechFlowDaily
X (Twitter):https://x.com/TechFlowPost
X (Twitter) EN:https://x.com/BlockFlow_News
