Author: Gu Yu, ChainCatcher
Earlier this month, the veteran crypto venture capital firm Variant announced the completion of a new $222 million fund, expanding its investment theme from the previous "digital ownership" to "autonomy."
This may seem like a typical fundraising event, but the underlying signal is anything but ordinary.
Variant partner Jesse Walden stated that in the future, the label of "crypto investor" might gradually fade away, becoming akin to that of an "internet investor." In other words, crypto is no longer an independent, closed investment vertical but rather a foundational technological paradigm embedded into mainstream sectors like AI, finance, social networking, robotics, data, content, and consumer products.
This is perhaps the most pragmatic answer crypto VCs have given in response to the AI onslaught: not competing with AI for narrative dominance, but attempting to become the underlying financial rails of the AI world.
I. Crypto VCs Begin to Blur Their Boundaries
In recent years, the fundraising logic of crypto VCs has been largely built on one premise: blockchain will give rise to a new generation of platforms, protocols, and applications independent of the Web2 world.
This logic was once highly persuasive as narratives like DeFi, NFTs, GameFi, Layer1, Layer2, modularity, restaking, DePin, and RWA emerged in succession. Funds that entered new narratives early enough could potentially reap returns far exceeding those of traditional equity investments through secondary market token liquidity.
However, this logic is now failing. The core reason is the significant weakening of the wealth effect within the crypto market itself. Bitcoin has retreated sharply this year, with several market views citing capital outflow from crypto ETFs, macro liquidity pressures, and investors shifting towards AI and major tech IPOs as key reasons. Meanwhile, AI and hard tech companies like SpaceX, OpenAI, and Anthropic continue to capture the attention of LPs and secondary markets, significantly diminishing the scarcity of crypto assets in terms of "growth stories."
This means crypto funds are not just competing with other crypto funds; they are competing with all assets representing future growth. AI, robotics, space, defense tech, and energy infrastructure are all vying for the same pool of LP risk capital.
In this context, "crypto-only" is transitioning from a professional label to a potential constraint.
If LPs believe AI is the most important technological variable of the next decade, a fund can hardly justify its irreplaceability merely by saying, "We understand tokenomics better." Especially in past cycles, many crypto projects failed to prove real revenue, user retention, and application scenarios, leaving behind structural issues like high FDV, low circulation, airdrop farming, and on-chain zombie applications.
This is why more and more crypto VCs are proactively blurring their boundaries.
YZi Labs has expanded its investment scope to three major directions: Web3, AI, and biotech, and participated in the $52 million funding round for AI industrial robotics company RoboForce this year.
According to a *Wall Street Journal* report in February this year, Paradigm is seeking to raise up to approximately $1.5 billion for its next fund, expanding its investment scope from crypto to "frontier technology" like AI and robotics, while still maintaining its crypto focus. In May, AI manufacturing company SendCutSend completed a $110 million funding round with Paradigm's participation.
In May, Haun Ventures announced the completion of a new $1 billion fund, expanding its investment scope to the AI agent domain. Founder Katie Haun stated that artificial intelligence will "increasingly conduct economic activity on our behalf," and services need to adapt accordingly for that future.
II. AI Agents Could Be Crypto's True Mass-Adoption Application
In the past, crypto projects often tried to get users to adopt products for the sake of "decentralization," but reality has proven that the vast majority of users won't change their behavior based on ideology.
Now, the crypto industry finds itself in an awkward situation: it still possesses unique capabilities like global reach, open finance, composability, asset issuance, and censorship resistance, but these capabilities have long lacked truly high-frequency, essential, and large-scale application entry points.
What is more likely to happen in the future is that users won't even know they are using crypto, but AI Agents, robots, financial applications, games, or content platforms will be calling upon stablecoins, wallets, smart contracts, and on-chain identities in the background.
According to Variant, autonomy is not merely automation. Automation solves whether machines can complete tasks for people, while autonomy focuses on whether users truly control their assets, identity, data, and decision-making power. Variant stated in an article that building autonomous systems requires solving a series of problems, including incentive mechanisms in adversarial markets, legal, governance, security, verification, policy, and geopolitical interfaces, with digital ownership being a key pillar of autonomy.
"The ideas that fueled the Web3 movement will find new momentum in the age of AI. We did a lot of experiments where crypto wanted to be seen as the product itself. But ultimately, we realized that crypto is the rails supporting many products, and its growth story has only just begun," said Jesse Walden.
This is perhaps the most important cognitive correction for the crypto industry in recent years.
Crypto doesn't necessarily have to be the front-end application users open every day; it can become the economic settlement layer between machines and machines, humans and machines, and applications and applications in the AI era.
If AI Agents are to perform tasks on behalf of users, they need wallets. If they are to autonomously purchase APIs, call computational power, pay for data, and subscribe to services, they need a low-cost, global, programmable payment network. If they need to carry identity, reputation, and assets across multiple platforms, they need an open account system. If external entities are to trust the results of their actions, they need verification and auditing mechanisms.
These are precisely the areas where crypto has accumulated capabilities over the past decade.
III. The Case of Tether's Investment
Crypto giant Tether's investment in NEURA Robotics is a representative case of this trend.
On June 10th, German robotics company NEURA Robotics completed a $1.4 billion funding round, with investors including Tether, Amazon, Nvidia, Qualcomm, Bosch, Schaeffler, and the European Investment Bank. NEURA stated that the funds will be used to scale up the commercialization of cognitive robots and humanoid robots, with plans to produce millions of robots by 2030. The company also disclosed that its order backlog already exceeds $10 billion.
On the surface, this is an investment in AI robotics. But for Tether, it's clearly not just a financial bet.
According to related press release information, NEURA's robotics platform is expected to integrate Tether's Wallet Development Kit (WDK), embedding self-custody wallet functionality directly into the robotic systems. This means robots may in the future receive micropayments for completing tasks, conduct transactions with other systems, or execute economic activities within pre-set human parameters.
This is one of the most imaginative new scenarios for stablecoins.
In the past, the largest users of stablecoins were traders, cross-border payment users, grey market arbitrageurs, and some residents of emerging markets. They solved the problems of transfers, settlements, and value storage between humans. But if AI Agents and robots begin to become economic actors, the frequency and scenarios for stablecoin use could be significantly amplified.
A robot can accept orders, complete transport, and receive micropayments in USDT; an AI Agent can automatically purchase data, call models, and pay for SaaS services; an automated supply chain system can automatically settle upon arrival, sensor verification, and contract confirmation. Compared to traditional banking systems, on-chain payments are naturally suited for this kind of high-frequency, low-value, cross-border, machine-readable economic activity.
This is also why AI is not just a competitive threat to crypto. AI may have captured crypto's capital attention, but it may also create the real demand that crypto has always lacked.
IV. AI + Crypto Is Not a Magic Formula
Of course, AI + Crypto does not naturally hold true.
Over the past two years, the market has seen too many crudely stitched-together projects: connecting ChatGPT to a Telegram group and calling it an AI Agent; wrapping a model API call with a token economy; stuffing data labeling, compute leasing, and agent platforms into a whitepaper. The problems with many such projects are no different in essence from those of the previous GameFi and SocialFi cycles: grand concepts, minimal revenue, heavy token emphasis, and light products.
Truly valuable AI + Crypto projects should meet at least one condition: they cannot exist without crypto, or they are significantly better with crypto.
For example, Agents need self-custody wallets and permission management; AI-generated content requires verifiable provenance and ownership; markets for models, compute, and data need open settlement and incentive mechanisms; the robot economy needs machine-readable payment networks; and autonomous organizations need transparent governance and enforceable rules. In these scenarios, crypto is not a marketing label stuck on the outside but a foundational component required for the system to function.
This is also the question that crypto projects and VCs need to answer next.
If they simply change their fund landing page to 'AI + Crypto' because AI makes fundraising easier, it won't change the industry's predicament. The market will eventually realize that most so-called integration projects lack both AI moats and crypto necessity.
But if they can identify the genuine convergence points among AI Agents, robotics, data markets, financial automation, and on-chain identity, the crypto industry may indeed usher in a new application cycle.
This time, growth might not come from more retail investors rushing into exchanges to buy new tokens, but from more machines, applications, and enterprises using on-chain rails in the background.
V. Conclusion
Faced with the impact of AI, the answer from crypto VCs is now clear: stop treating crypto as an isolated vertical, but reinterpret it within the larger wave of technological progress.
This is both proactive evolution and a forced pivot.
When LP capital flows to AI, when entrepreneurs' attention flows to AI, when secondary market risk appetite flows to AI, crypto funds that continue to talk only about Layer1, DeFi, NFTs, blockchain games, and airdrop-led growth will find their space increasingly narrow.
But this doesn't mean the crypto story is over. On the contrary, if AI Agents truly become the new internet users, if robots truly become new economic participants, and if automated systems genuinely start executing more and more transactions on behalf of humans, then the wallets, stablecoins, smart contracts, on-chain identities, and open finance networks built over many years by the crypto space might finally encounter high-frequency, essential, and non-speculative usage scenarios for the first time.
The crypto industry needs a new narrative more than ever, but it needs new, real demand even more.
