The End of the Speculative Era: The Crypto Market is Being Rewritten by AI

robot
Abstract generation in progress

Author: Kerman Kohli Translation: Shan Ouba, Jinse Finance

I have been contemplating the next direction of the crypto market and the current stage we are in. The entire market is now generally mired in confusion and lost direction, as if a temporary dilemma on the way to a great journey. The era of easy profits in the market has already ended.

Almost all tokens have plummeted across the board, which also means that industry-wide profits are beginning to collapse comprehensively. Essentially, all Ponzi economic models have vanished, and there is no truly speculative space left in the market.

Indeed, the crypto space still has applications related to payments and financial technology, which can help traditional finance improve efficiency by another 25 basis points, but that is not the original intention of our involvement. Those who still remain in the crypto industry are idealists eager to witness disruptive systemic changes rather than being co-opted by capital elites.

I often pay attention to industry culture, and currently, many crypto projects’ cultural core has already become subordinate to capital. This also tells me that this path has gone astray. Therefore, we must shift our focus to the fringes of the industry, exploring areas that the public and capital elites have not yet truly noticed or fully understood — and I am convinced that this area is artificial intelligence.

I don’t want you to dismiss this as just another hype piece promoting low-quality decentralized physical networks, decentralized reasoning, or AI + crypto concepts. Next, let’s explore this with rigorous logic.

Machine Economy

When I previously discussed investing in AI stocks, I introduced this concept. Now I hope you can temporarily set aside your preconceived notions and unleash more imagination.

My intuition tells me that a brand-new parallel economic system is emerging, with operational efficiency far surpassing that of the familiar human economy. The efficiency of human economic operation is painfully slow, requiring coordination among countless individuals with conflicting interests. What would it look like if a completely new economic form was born, led by machines and taking on more and more tasks?

This economic system even has its own inflation characteristics. The costs of most cloud computing platforms are increasing at over 10% annually. If your computational application scenarios lack sufficient value, you will be eliminated by the market, and your computing resources will flow to more valuable fields. Computing power is the scarce resource of this new world; the more efficiently it is deployed, the greater its usable value.

With careful observation, many clues can be found at the industry fringes:

  • After people started using ChatGPT, the competition is about how much time they spend daily interacting with prompts

  • After intelligent coding agents went live, people began depleting their daily token quotas at higher frequencies and showing off their usage

  • Tech labs launched highly popular $200 monthly subscription plans, with model companies happily subsidizing them

  • As the capabilities of intelligent agents continue to improve, these plans are becoming unaffordable for many

  • Data centers planned for launch failed to come into operation on schedule

  • High-performance chips are becoming increasingly scarce

  • Many service providers’ computing power is in short supply, and prices are rising

  • End users are starting to purchase hardware themselves to obtain computing power

While improving the efficiency of computing power usage, people also hope to extract more value from it.

Ensuring that intelligent agents complete their designated tasks during operation is becoming increasingly important, including ensuring they make correct tool calls.

It is crucial to enable intelligent agents to perform more operations without manual authorization. And these tasks will initially start with financial transactions.

Everything will begin with API payments, likely through small transactions, enabling rapid trial of API services. You just need to give a command: “Register all 10 API service providers you think can solve my problems, and select the best fit for my tech stack,” and the intelligent agent can quickly register and pay for these 10 providers, testing in parallel to find the optimal solution.

As development progresses, these services will become more complex, and the funds earned by intelligent agents will require more sophisticated financial services like lending. We can already see the embryonic form of this trend, such as USDai bringing GPU financing on-chain (I find this attempt interesting, but it’s far from the complete form we expect).

So, the question is: when will all this arrive?

In my view, ordinary users can now experiment with frameworks like OpenClaw. Currently, intelligent agents are mainly used to improve convenience and optimize workflows, but over time, they will be optimized to generate more economic value. When permission divisions become more refined, control tools continue to iterate, and 24/7 operational intelligent agents become a reality (most agents today can only perform tasks lasting tens of seconds to 10 minutes), we will witness the rise of proactive value trading by intelligent agents. I believe this milestone will arrive within 12 months at the earliest, and no later than 18 months.

A New Crypto World

This is the part that excites me the most. I am fundamentally a technologist, working in the crypto industry, but I often feel that I am dealing with a group of lazy-minded finance practitioners who are neither top-tier financial talents nor excellent technologists — a poor combination of both. I hold great respect for cryptopunks; they integrate expertise from multiple fields and are true interdisciplinary talents.

When intelligent agents enter the crypto space, they won’t care about narratives driven by short-term traders or hype; they will focus solely on value and tasks to be completed, after all, they rely on expensive computational power. In this new world: technical strength is paramount, product functionality is critical, rigorous system design is essential, and adversarial thinking is vital. All qualities that give the crypto industry depth of thought and an elitist ethos will return.

This entirely new crypto world will no longer be a game of venture capital issuing high-market-cap tokens to harvest retail investors, but an era where data is king. This leads to my next point:

Data in the crypto industry will undergo a thorough revaluation in this transformation. Previously, crypto data was always a secondary role, industry standards lagged, and few related companies existed, many of which are now disappearing. This is understandable: data previously only served humans, and while data analysis was valuable, it was hard to directly translate into tangible profits.

But in the new world, data is the lifeblood of on-chain transactions involving intelligent agents. Massive numbers of intelligent agents will generate enormous on-chain data, and they will also read data from other agents, leading to explosive growth in data demand.

I have another article that elaborates on the current difficulties faced by data companies due to computing power shortages, so I won’t go into detail here.

Besides data, other sectors will also undergo disruptive revaluation: custodial services (programmable custody for intelligent agents, not traditional models like Fireblocks), oracles (the importance of data feeds will far surpass today’s), trustworthy decentralized attack-resistant code for intelligent agents, and a whole new asset class born from the integration of intelligent agents and the crypto industry.

This is the crypto new world I look forward to. We just need to clear out the dross in the industry, and its outline will become clearer.

x402 is a good start, and it’s gratifying that it is under the Linux Foundation’s stewardship rather than exclusively controlled by Coinbase.

Conclusion

We are about to embark on a brand-new journey, and I am patiently waiting for this era to arrive. If you feel frustrated with the current state of the crypto industry or confused about the future, I hope this article can light a beacon for you. Remember, all infrastructure must be in place before the intelligent agent economy begins to conduct financial value transactions. Prepare early, start planning, and when this trend becomes clear, acting late will be too late. The crypto industry’s funding will heat up again (like in 2020), market competition will intensify (few new entrants are joining now), and winners will quickly stand out. The crypto industry has always been a winner-takes-all market, with only 2 to 3 winners in each sector. You must have a clear understanding of this future world and dedicate yourself to building it.

There are two teams I want to highlight in particular, as I believe they exemplify these principles perfectly: salt.space (of which I am an angel investor) and herd.eco (an excellent team).

View Original
This page may contain third-party content, which is provided for information purposes only (not representations/warranties) and should not be considered as an endorsement of its views by Gate, nor as financial or professional advice. See Disclaimer for details.
  • Reward
  • Comment
  • Repost
  • Share
Comment
Add a comment
Add a comment
No comments
  • Pin