Web3 is sick, but the cure isn't AI.

Author: Zhou, ChainCatcher

Over the past year, breakthroughs in AI technology have far exceeded expectations. From GPT-4o to the emergence of various AI agent tools, new capabilities are being pushed to new limits every few months.

The explosive popularity of next-generation AI agent products such as OpenClaw has made many crypto professionals realize for the first time that AI has begun to touch the boundaries of their work.

As a result, an "AI anxiety" has spread throughout the crypto community.

From macro-strategies to micro-work methods, from leading institutions to frontline practitioners, many crypto projects and practitioners have begun to re-examine their own value and work models—will what they do be replaced? Does the industry still have a future? These questions have been repeatedly raised over the past year.

First, the project owners and organizations panicked.

Product-related responses

When AI agents can directly take over the keyboard and mouse and perform tasks on a computer, users will be able to complete on-chain operations entirely through AI in the future, without needing to open exchanges or wallets. So, where will the value of exchanges as gateways lie?

Earlier this month, leading exchanges such as OKX, Binance, and Coinbase announced their AI product development plans.

Among them, OKX launched the Onchain OS AI upgrade, supporting AI Agents to operate autonomously across more than 60 blockchains and more than 500 decentralized exchanges; Binance announced that it will provide each AI Agent with Binance-grade brain, directly embedding exchange-level trading intelligence into the AI ​​Agent; and Coinbase launched Agentic Wallet for AI autonomous operation.

The logic behind these actions is simple: users can operate with AI, but as long as the infrastructure is still here, the entry point is still here.

However, despite the excitement, the actual data is much more sobering.

The a16z report mentioned that previously reported data claimed that AI agents completed $24 million in payments within 30 days, but after independent verification and elimination of fraudulent transactions, the actual figure was approximately $1.6 million, a reduction of nearly 15 times.

This indicates that the scale of on-chain payments by AI Agents is currently far smaller than what is being portrayed by the outside world; rather, emotions and narratives are outpacing actual data.

In an interview with ChainCatcher, David Gan, founder of Inception Capital, stated that not all AI needs crypto. Today, most AI applications are essentially still engaged in information processing, content generation, and workflow efficiency. However, once AI agents move beyond simply talking to actually doing things, starting to call services, manage budgets, initiate transactions, and complete settlements, crypto will gradually transform from an option into infrastructure.

Staffing reduction

In February of this year, fintech company Block laid off approximately 4,000 employees. CEO Jack Dorsey had previously admitted that the layoff decision might have been a mistake, stating that the rapid development of AI technology prompted the company to restructure its 6,000-person team. However, the company quickly contradicted itself, quietly recalling some employees a month later due to paperwork errors and a shortage of staff in infrastructure positions.

The crypto industry itself has also been affected. Recently, Crypto.com announced it would cut about 12% of its jobs to advance enterprise-level AI integration; Gemini has reduced its workforce by about 30% since layoffs at the beginning of the year and is introducing AI tools to improve productivity; the Algorand Foundation announced it would cut about 25% of its staff, and Messari also announced staff reductions.

The trend of AI replacing traditional technologies is no longer limited to the tech world; it has truly impacted the crypto industry. Forced adoption of AI tools, reduced hiring, and a reassessment of job necessity are all being addressed.

However, there is also a more objective view. Crypto practitioner Forest Bai told ChainCatcher that layoffs are a result of the combined effects of the bear market and the use of AI. For business owners, when the cost-benefit ratio of AI is lower than that of humans, replacement will occur. This logic existed even before the advent of LLM.

II. The anxieties of individual practitioners are hidden deeper.

Exchanges can mask their anxiety with product development, but for individual practitioners, it seems that not using AI tools now means being out of touch with the times.

CZ tweeted that after installing OpenClaw, he didn't have to do anything else, and all his time afterward was spent adjusting that useless OpenClaw.

This self-deprecating remark resonated widely. One crypto builder bluntly stated that people without a technical background have virtually no chance of playing with OpenClaw, as the debugging difficulty is extremely high.

For ordinary practitioners, the sheer barrier to entry—just configuring the required data access and skill modules—is enough to deter most. The hype is real, but truly viable applications are currently limited to a select few.

The deeper anxiety comes from those who have already left.

Over the past two years, a group of the smartest and most restless people in the crypto community have emerged in the world of AI.

• Alex Atallah, co-founder of OpenSea, founded OpenRouter, directly applying his integration thinking accumulated in the crypto industry to AI infrastructure.
• Leopold Aschenbrenner, a core member of the FTX Future Fund, now manages a multi-billion dollar AI investment fund that focuses on power infrastructure, semiconductors, and computing centers.
• Avital Balwit, who also came from FTX Future Fund, is now the chief of staff to Anthropic CEO Dario Amodei, participating in the company's top-level strategic decisions.

These people left not because encryption failed. Or rather, encryption served as their training camp, giving them a keen sense of risk, sensitivity to power structures, and the ability to make judgments in highly uncertain environments. They took these abilities to a larger battlefield.

For those who remain, the anxiety isn't so much about being eliminated by AI, but rather a more helpless predicament. Renowned crypto KOL 0xSun frankly admits that for most ordinary people, the booming development of the AI ​​industry has little to do with whether they can enjoy its benefits. Most people, due to limitations in mathematical ability, education, and upbringing, are simply unable to truly enter the AI ​​industry.

Forest Bai points out that the current anxiety among crypto professionals stems more from the continued decline in market prices than from the disruptive nature of new technologies. Another crypto KOL suggests that in the AI ​​era, core human capabilities will shift from productivity to decision-making. AI won't replace people, but rather the standardized, replicable, and automated aspects of human beings. Those who remain may have the opportunity to redefine their place in the industry.

Third, the narrative dilemma is real, but AI is not the solution.

The entire crypto community's reaction to AI shares a common premise: crypto needs AI to save its narrative. This logic is understandable, but it doesn't stand up to scrutiny.

Using new narratives to cover up old problems

The narrative dilemma facing the crypto industry in the past two years is real. Web3 has a grand vision, but the number of users it retains remains limited; DeFi's transformation of the traditional financial industry is still ongoing; the NFT creator economy bubble has burst; and the metaverse has long since died out.

After every narrative fizzles out, the industry needs to find new stories to fill the void, and this time it's AI's turn.

David Gan points out that many projects make a typical mistake: they assume that AI and crypto must be combined, and then try to find scenarios for this combination. Truly good projects should demonstrate that removing either component significantly reduces the product's value; this is a very high barrier to entry.

In an interview with ChainCatcher, David and Daniil Liberman, co-founders of the decentralized AI computing network Gonka, also stated that most projects attempt to integrate AI at the narrative or token economic level, rather than at the infrastructure level. The real challenge is not adding AI to crypto, but building a system that can provide open and scalable access to computing resources.

In an interview with ChainCatcher, Chris Feng, CEO of AI infrastructure Axis Robotics, added that labeling products with AI is essentially an attempt to capitalize on emotional trends rather than solving real problems. The key dividing line is whether you are solving real-world problems.

The real dilemma of the crypto industry has never been a lack of compelling narratives, but rather a lack of real-world applications, an inability to retain users, and a failure to generate sustainable business value.

AI won't change this. Exchanges adding AI features won't make more people need crypto; projects labeling themselves with AI won't actually increase on-chain activity.

Where does the real intersection of AI and Crypto lie?

Of course, some people are seriously considering the real intersection between AI and Crypto.

After AI agents move from dialogue to execution, they begin to require autonomous payments and on-chain settlements. The account system and KYC process of the traditional financial system are not suitable for this, and stablecoins and programmable contracts have a real place in this scenario.

Chris Feng believes that agents need autonomous payment capabilities, verifiable identities, and traceable operation logs, which are not naturally addressed in traditional Web2 systems. Crypto fills this gap perfectly.

This trend is evident in the with the growth in demand for AI-assisted payments, the supply of USDC has rebounded to near its historical peak, and adjusted transaction volume has increased by over 90% year-on-year, increasingly being used in scenarios beyond crypto transactions. Meanwhile, Stripe, Cloudflare, and Google have also embedded their on-chain payment standards into their respective AI-assisted payment protocols.

In an interview with ChainCatcher, Bill Sun, a researcher who has long been involved in the field of AI, stated frankly that Circle's real competitor should be Stripe, not USDT. He believes the quasi-financial layer needs companies like Stripe that take strategic value seriously and create truly valuable infrastructure. He also mentioned that the crypto industry has focused too much on short-term retail trading and hype in recent years, wasting a significant amount of meaningful development energy that should have been used for truly valuable things.

On-chain payments are a real direction, but they may be solving AI-related problems rather than cryptographic ones. As Forest Bai stated, while he acknowledges stablecoins as a potential path for AI payments, their current share is very small. Stablecoins and blockchain payment networks are existing infrastructures, and there's no need for entirely new solutions.

David Gan stated that the value of AI to Crypto lies not in giving the industry another hot label, but in whether it has the opportunity to enable blockchain to truly serve software, machines, and automated execution.

If the industry can seize this opportunity to truly delve into the underlying mechanisms of the machine economy and open network collaboration, it presents a structural opportunity; however, if it merely treats AI as a new narrative applied to old models, it will quickly cool down, just like the previous rounds of narratives.

Conclusion

Ultimately, the AI ​​anxiety within the crypto community is a stress response to external shocks to an industry whose narrative has been overstretched. The anxiety is real, but AI is not the cure.

Over the past few years, after every major narrative collapsed, the crypto industry has healed itself in the same way: finding new stories, creating new concepts, and attracting a new wave of people.

This time is no exception. Exchanges are launching AI features one after another, practitioners are rushing to evaluate lobsters, and project teams are vying to label them. Behind the excitement, however, there is still no answer: what problems can the on-chain world actually solve for real users?

This is a question that AI probably cannot answer.