Every time the market goes down, the same scene always appears:

A group of people jumps out and says, "See, I told you it was a scam";

Then when the market turns, the same group lines up, chasing highs and shouting bullish.

This extreme back-and-forth of emotions has numbed me after eight years in this industry.

But the fact is:

When we talk about price, we are actually talking about the future. Because price is never "now," but rather the market's discounting of the future. If we continue to focus only on price, the future will slip away from us.

Moreover, there is a very real pattern in bear markets: everyone sees the decline; I see the stratification. Emotions are retreating, the industry is filtering, and the future is being rearranged.

At the same time, I am also thinking, "Why are there more and more new projects, but truly meaningful things are becoming fewer and fewer?" "We all know the industry is going to change, but how exactly will it change?"

That's why I want to write this article. Not to pump up enthusiasm, nor to shout "the bull market will return," but to show you in the simplest and most genuine way—what is "currently happening in this industry" through Jiayi's eyes, and "what will happen in the future."

The bull market is noise; the bear market is a magnifying glass, a microscope, a truth-revealing lens. In the fog of emotions, I want to take you to a higher vantage point to see where the future's path lies after the tide recedes.

1. Why I say the bubble is the biggest boon in 2025

In 2025, there is a very obvious fact that most people are unwilling to admit: we did not wait for the "copycat bull," but instead welcomed a systemic deflation of bubbles ahead of time.

Many people interpret this as a bad thing, but if you extend the timeline, you will find that—these years of "killing before rising" are actually the best period for the industry to develop a truly mature structure.

Why? Because everything is exactly like the early days of the internet, only the speed has been amplified by crypto's leverage.

If you want to truly understand today's crypto, the simplest way is to—view it as an "accelerated version" of the early internet.

Many people think that the chaos, bubbles, and speculation in crypto are unique "original sins" of this industry.
But if you extend the timeline, you will find that this is not a special case at all, but rather a standard feature of almost all technological revolutions—

The internet in 2000 was just as crazy.

Previously, I wrote an article titled “From 'Burning Money' to Building Industry Ecosystems: Web3 is on the Same Path as the Internet”, which discusses this logic: under the premise of amplified leverage efficiency, the market operation methods of Web2, the internet, and crypto are essentially converging. It's just that what took the internet twenty years to accomplish, crypto may wrap up in less than ten.

1.1 Internet finance went through the same "incomprehensible → frenzy → collapse → reconstruction"

If you think crypto's volatility is dramatic, it's just because you forgot the first half of the internet's life.

In 1999, as long as a company had ".com" in its name, it could raise funds. eToys skyrocketed 900% on its first day of trading, and investors were as crazy as during the early days of crypto's copycat season.

Then the bubble burst.

The Nasdaq fell from over 5000 points to 1114 points; media headlines screamed "internet scam"; everyone said the internet was doomed.

That emotional period was almost identical to today's crypto:

• Those who don't understand it call it a scam

• Those who do understand it are also intoxicated by the bubble

• In the end, everyone gets trampled

• Then everyone collectively doubts the future itself

But ironically—the real internet era began the moment the bubble burst. When the tide recedes, we not only know who was swimming naked. It also helps us filter out the true players most likely to swim to the other shore.

In 2002, Amazon's stock price was only $0.6. Google had not yet gone public. The companies that were left behind built the real infrastructure, business models, and profit methods during those years when they were least visible.

Crypto now resembles 2002-2004: not the hottest years, but the most critical years.

The bubble burst cleared away the noise, and trends, directions, infrastructure, and real players began to enter the "construction phase."

This is why I say: the bubble in 2025 is a boon. Because truly important things will only emerge after the bubble.

1.2 Denial of denial: the industry does not rise in a straight line, but spirals upward

I particularly like the concept of "denial of denial" because it perfectly describes the evolution of technology and industries.

"Denial of denial": the development of things is never a straight upward line. It is more like "going up in circles": every time you think you have returned to the starting point, you are actually at a higher level.

The most typical example is the three iterations of computing architecture.

• 1950s: IBM Mainframe ------ Centralized
  Computing power was concentrated in a few institutions—governments, banks, large enterprises. This was the first concentration: whoever has the machines has the power.

• 1980s: PC Revolution ------ Decentralized
  Steve Jobs said, "Computers are the people's bicycles." Computing power moved from data centers to everyone's desktops, and everyone owned their own computers, which was a denial of "centralized power." Computing began to decentralize, becoming personal and localized.

• 2010s: Cloud Computing ------ The "denial of denial" of centralization, standing at a higher level
  AWS, Alibaba Cloud, and various public clouds re-concentrated computing power back to data centers.
  On the surface, this looks like a return to the "mainframe era":
  Once again, a few giants control massive computing power.

It seems that cloud computing has returned to "mainframe centralization," but it is fundamentally not the same level:

• Terminal sovereignty comes from PCs,
• Elastic computing power comes from the cloud,
• The advantages of two generations of technology are integrated.

Crypto and internet finance are currently experiencing the exact same path:

• First phase: no one understands, but the direction is vaguely there

• Second phase: everyone is fervent, the bubble inflates to the sky (we just went through this)

• Third phase: bubble bursts → removing the false to retain the true → spiraling upward (this is 2025)

Do you think the industry has "returned"? In fact, it has not returned to the starting point, but to a higher "starting point."

Trends, tech stacks, capital structures, and regulatory paths are all being rearranged in this round.

2025 will be the most important signal before the next round begins.

2. 2025 is the prologue, 2026 is the main text: the true trends of the industry are taking shape.

Whoever solves the directional issue of will own the world for the next five years.

The feeling of 2025 is actually quite strange.

The K-line looks like a bull market, but the sentiment feels like a bear market;
Regulation is advancing, but expectations are declining;
Narratives are flying everywhere, but making money is harder than any other year.

In the future, I boldly predict that pure Crypto Native innovations will face bottlenecks. The real breakthroughs and effective innovations will emerge in the direction of 'Crypto ✖️ Traditional Finance's complete integration,' which is the compound innovation that can simultaneously address the financial market needs of both Web2 and Web3.

2.1 Everything happening in 2025 points to one thing: the industry is "restructuring."

Bitcoin surged to $126,000, mainstream assets followed suit, but altcoins were increasingly lukewarm. The secondary market seems to be rising, but the final account balance hasn't increased… This is the most surreal experience of 2025.

But when you look at 2025 from a "structural" perspective, this year suddenly becomes very clear: This is the first time that the U.S., Europe, and Asia are all accelerating policy changes in the same year:

• For the first time, the U.S. SEC's attitude towards ETFs and crypto has become friendly

• BTC, ETH, SOL, and XRP have all launched spot ETFs. The door is open

• The stablecoin bill has provided a clear framework for the first time nationwide

• Europe's MiCA has truly landed, with dozens of licensed institutions emerging

• RWA has become a key pilot direction for regulators everywhere

Crypto is being institutionalized into the global financial system for the first time. The earliest OGs in crypto hoped for it to enter mainstream finance. Do you remember the title of our article? When we talk about price, we are actually talking about the future. The market's resurgence requires new stories for redemption, and I personally believe it is "Crypto ✖️ Traditional Finance's complete integration."

2.2 On the track level: high heat, weak prices, this is a typical "verification period"

RWA, AI, stablecoins, L1 (lasted less than a month), prediction markets, perpetual contracts, on-chain asset management, DAT… every term can ignite a short-term sentiment. Emotions aside, prices are not rising. Short-term excitement is replaced by a longer-term secondary market slump.

📒2025's RWA: The underlying infrastructure is complete, and the regulatory framework is beginning to take shape, providing "implementable" conditions for the development of RWAFi.
Let’s hype Plume's contribution to the implementability of RWA. I mention Plume not because I invested in it, and certainly not because its secondary market performance has been mediocre recently. However, I firmly believe that PLUME best represents the "infrastructure moment" that RWA has experienced this year.

First of all, Plume allows real-world assets to:

• Legally enter the on-chain yield system,

• Be participated in by ordinary users (not just institutions),

• Achieve on-chain distribution (collaborating with traditional financial brokers)

• Be DeFi-ified, composable, and liquid.

Just a few developments from the past two months: Securitize's assets have been introduced into the on-chain yield system; products from Apollo, VanEck, and BlackRock can be used in on-chain strategies; Bluprynt's KYI brings "issuer transparency" on-chain; it has become a registered transfer agent with the SEC, connecting with the regulatory-level grassroots interface of the SEC/DTCC.
These things may sound dull, and users may not know how to engage with the secondary market's FOMO. But essentially, you only need one sentence:

As the world's first public chain for RWA Fi, Plume has essentially flattened the mountain in front of RWA and paved the "road" needed for RWA to land.

Because of teams like Plume building the framework for RWA in 2025, more excellent assets under the RWA-Fi system will have the opportunity to truly take off in 2026.

3. AI × Crypto: There are not many truly valuable parts, but the value density is extremely high

To put it bluntly, I am afraid of offending people by saying that the AI track looks lively, but there are actually very few truly valuable things.

3.1 Too many false propositions

Take the grand narrative of LLM, which has made the market FOMO, as an example. I believe traditional AI's large language models have matured to the point where "crypto simply cannot compete."

ChatGPT, Claude, Gemini, DeepSeek—these models are backed by real global capital, built on tens of billions of dollars' worth of computing power, data, distributed training systems, and engineering teams.

Does Web3 want to "create another LLM"? It's not that the dream is impossible; it's that physical laws do not allow it. And what is it that traditional AI's large language models cannot satisfy you with? Web3 does not need to reinvent the wheel. I suggest that projects of this type stop burning cash on technology and quickly seek real demand scenarios as a way out. After all, 2025 has another typical characteristic: users are harder to deceive. 😁

I hope one example helps everyone understand why most AI projects in 2025 easily become "hollow" as they go along.

3.2 Building the foundation for compound innovations that meet the financial market needs of both Web2 and Web3

First direction: Building a "value incentive and collaboration structure" for AI
Focusing on on-chain incentive mechanisms that traditional AI lacks, a globally settlable economic system, and a traceable contribution network, after all, models, data, computing power, and agents all require incentives, collaboration, and distribution. This is the problem Sahara AI aims to solve.
Second direction: The economic and execution systems of agents ⚠️ The true strongest intersection of Web3 and AI.
Still focusing on better solving the demand for automated decision-making, execution, trading, and settlement, based on the limitations of agent capabilities in the Web2 world.

In the Web2 world, agents' capabilities are currently constrained by:

• Inability to make autonomous payments

• Inability to manage assets

• Inability to call across systems

• Inability to execute tasks without permission

• Inability to transparently track behavior

Whereas in the crypto world, these are all native capabilities:

• Wallets

• Smart contracts

• DeFi strategies

• On-chain identities

• Stablecoin settlements

This is the direction that most Web3 AI projects are currently building: building the underlying "execution system" for the future Agent era. Starting with financial-related executions is clearly the most urgent and largest market demand.

3.3 The third direction: AI payments, the biggest disruption in the next five years

If I had to summarize the super trend of the next five years in one sentence, it would be the revolution of payment methods driven by AI agents. When agents start executing all instructions from users to application terminals, such as selecting, placing orders, asset allocation, and strategy management, payment will no longer be the endpoint of user actions, but the "underlying capability" of agent operations.

In such a world, security, verifiability, global availability, extremely low cost, second-level settlements, and 24/7 programmable cash flow will all be essential conditions for AI agents.

Traditional payment systems cannot achieve this. But with AI combined with stablecoins and on-chain identity systems, everything becomes feasible. In the future, you will see more and more AI applications begin to let agents

• Automatically call stablecoins

• Create multi-signatures

• Set up custody and strategies

• Manage on-chain identities

Seamlessly integrating value transfer itself like calling an API.

Even giants like PayPal, which have been in traditional payments for decades, are starting to accelerate their layout, investing in the next generation of projects with "AI-native payment capabilities," such as Kite AI. Many people will say: Kite isn't really providing payment technology services to AI agents on a large scale yet, is it just hot air?

Let’s return to the title, "When we talk about price, we are actually talking about the future." The real question should never be "How many agents does it serve today?" but rather in the AI-led future economic model:

Who is already building the foundational capabilities needed for the future? Who is laying the groundwork for the value network of the AI era?

Just like after Coinbase released the 402 protocol, dozens of "cryptographic new projects" emerged within days. Bad coins will prematurely overdraw the market's expectations because when everyone discusses prices, they represent the future in their hearts. The efficiency of issuing tokens in the industry, along with the prosperity promoted by MEME and junk copycats, has made it even more difficult for entrepreneurs. However, good projects in the market are always scarce, and in this regard, raising the entrepreneurial threshold is not necessarily a bad thing.

4. Stablecoins: The most early-recognized but most worthy ALL-IN track in 2025

To be frank, if there is one track that should be "quietly but definitely taken seriously" in 2025, it is stablecoins.
I believe it is still in the early narrative stage, the market does not really understand it, and the vast majority of stablecoin-related projects that emerged in 2025 are neither stable nor applicable, and I believe the market has not yet entered a true FOMO stage.

4.1 The most eye-catching stablecoin event this year: the Trump family issued a coin

Yes, the Trump family clearly issued a stablecoin: World Liberty / World Finance. The less content, the bigger the event.

4.2 The real demand scenarios for stablecoins in 2025 are quietly accelerating

① Government bond reserves become mainstream → Compliance accelerates

All leading stablecoin issuers are basically moving towards transparency in reserves and compliance.

② On-chain payments have been "forced to upgrade"

Please refer to the previous section on AI payments.

On-chain payments are the infrastructure driven by the demand from the development trend of AI agents. Note that this market is not limited to the Web3 market. As I mentioned earlier, the future trend market will definitely be a larger market that integrates both Web2 and Web3.

Once on-chain payments mature, the ceiling for stablecoins will be directly raised.

③ Stablecoins are seeing "structural stratification" for the first time

Previously, stablecoins seemed to have only one logic: "Is it USDT or USDC?"

But starting in 2025, structural stratification has emerged:

• Centralized stablecoins (USDT, USDC) → Policy/institutional scenarios

• Exchange stablecoins (FDUSD) → Trading and new issuance scenarios within exchanges

• On-chain native stablecoins (DAI, USDL, etc.) → DeFi scenarios

• RWA stablecoins → Financial institution scenarios, on-chain settlements

• Payment stablecoins → AI agents, cross-border e-commerce scenarios

4.3 In the realm of stablecoins, the vast majority are "garbage projects eager to cash in on the hype"

Whenever a track begins to gain attention, a bunch of "pseudo-themed projects" will first emerge. To issue stablecoin concept tokens, teams naively believe they have thought of a genius scenario, and then the market starts mass-producing shells. A clean sweep of some TVL subsidy activities + packaging a yield scenario is essentially token overdraw, and since users love to farm, they just repeat the old path of btcfi with TVL, package it, and go to exchanges without caring about the future mess.

Users may indeed engage in short-term FOMO, but they will genuinely be disappointed with the entire stablecoin track due to these "pseudo" stablecoin projects in the medium term, while in the long term, the true giants will rise, and the secondary value of stablecoins is currently severely underestimated.

The core value of stablecoins has only two: stability; usefulness. If you can't even achieve "real reserves + consensus usage scenarios," issuing more white papers is meaningless. I guess someone here might say that USDT's reserve ratio is not 1:1. Sorry, but the consensus usage scenario of USDT is something that even the Trump family cannot surpass. To put it bluntly, the essence of currency is that as long as everyone believes it is stable, it is stable. The core advantage of USDT is that its first-mover advantage has brought about the earliest market consensus and years of operational experience in "maintaining stability." I stubbornly believe that this inherent advantage cannot be easily surpassed by a project through so-called stories to define an unstable asset.

4.4 The explosion of stablecoins will be larger, more stable, and longer-lasting than anyone imagines, and the stablecoin track will develop in a structural manner

• Policy recognition and strategic layouts by various governments

• Entry of financial institutions

• Various payment scenarios represented by AI agents, growth in cross-border payments, daily payments in L2, etc.

• Understanding of the lowest cost for new users to enter crypto; reserves of CEX; demand for OTC fiat currency

• Large-scale liquidation of RWA

• The medium for the complete integration of crypto and traditional finance

I tentatively define stablecoins as the "underlying fuel" for all future tracks.

There are many tracks that will perform well in 2025, as long as they focus on more efficient or diverse forms of financial investment. Gambling is human nature, and tracks are eternal. What matters is the experience. I won't elaborate further.

5. When we talk about price, we are actually talking about the future

How to use "the next five years" to reverse-engineer current opportunities? By this point, you may have already sensed one thing:

If you agree with me that the most important main line for the next five years is "the complete integration of crypto and traditional finance," then in the bear market, we should shift our focus from secondary prices to thinking about industrial value opportunities and pricing them. Just like the early MATIC (later renamed Polygon) in the L2 sector, which once broke below par, or SOL which fell to 9 dollars. In a bear market, look at price anxiety, and rejoice in future value opportunities.

From the market's perspective, what is price determining? It is determining emotions; it is determining the fluctuations of expectations; it is determining the heat of narratives.

But I believe the real value gap is determined by: who is laying the foundation for the framework of the next five years; who is building financial infrastructure that serves both Web2 and Web3; who is embedding themselves into "the financial system of the world under future liquidity integration," rather than just self-indulging on-chain.

So, if you agree with the logic I presented earlier,
then there is a very direct question:

👀 How should we view the current prices?

5.1 First, clarify a fundamental logic:

In the next five years, the vast majority of unicorns will not be "pure crypto-native giants."

"Pure crypto-native innovation" is carried out in the absence of a traditional financial market background, resulting in extreme on-chain innovation; it is also the crop driven by the lack of foundational blockchain or Web3 infrastructure. Over the past decade of blockchain history, we have experienced countless peaks:

• Native public chains

• Native DeFi Legos

• Native NFT & Game narratives

• Native DEX, derivatives, lending protocols

The giants from the last round (exchanges, public chains, leading DeFi protocols) have basically occupied the high ground of "on-chain native infrastructure."

Most "purely on-chain" things are either minor innovations, skin-deep changes, or regulatory evasion.

"The number of new projects is increasing, but truly meaningful things are becoming fewer."

Because the truly next layer of innovation must meet three conditions:

1. It can connect with Web2 and serve Web3

2. It can be used by real-world users, institutions, and funds

3. It can be embedded into the real financial system, rather than just cycling internally within the crypto space

In other words:

Whoever can truly connect crypto to "the flow of real-world money" and "the real financial system,"
will be qualified to obtain the largest valuation premium in the next round.

5.2 Reverse-engineering what is "worth watching" from the three main lines of the next five years

• Main line one: RWA-Fi → Turning real-world assets into on-chain composable production materials → Excellent real yield assets become globally accessible

• A large number of excellent assets with annualized returns of over 10% have not flowed into crypto, and there is currently a basic vacuum. Most of these assets have high user thresholds, and many people want to participate but are limited to institutions, large holders, and those with connections.

• U.S. and Hong Kong stock IPOs

• Real linkage between crypto and stocks, rather than being limited to pure DAT-based financing investment logic

• Main line two: AI Agents fully bloom in 2026------a true epoch-making turning point

Traditional AI companies have already pushed the "large language model competition" to its limits from 2023 to 2025. The era of models is over; the era of agents has begun. For crypto, the future breakthrough lies in—making agents economic, executable, and trustworthy.

• The "mass landing period" of AI agents, especially the formal surge in financial scenarios

• Web3 will undertake the most critical part of agent landing: the economic system will become a red sea

• Agent incentive systems

• On-chain collaboration systems

• Agent task markets

• On-chain economic models (yield, payment, custody)

• Efficient and trustworthy AI stablecoin payments will connect every agent task

• Main line three: Stablecoins & New Settlement Layer (the most difficult to make understandable)

Summary: Finally, let's revisit the core arguments of the article:

✍️ When we talk about price, we are actually talking about the future.
If you only focus on price, the future will definitely slip away from you.

✍️ The bull market is noise; the bear market is a magnifying glass, a microscope, a truth-revealing lens.

✍️ The bubble in 2025 is not a bad thing; it is the "fracture period" before the industry truly grows.
The cleaner the bubble bursts, the clearer the future becomes.

✍️ Pure crypto-native innovation is nearing its ceiling.
✍️ The narrative direction for the next five years in the industry: "the complete integration of crypto and traditional finance."

✍️ RWA-Fi, AI payments, and stablecoins are the three super main lines for the next five years. They are not trends; they are foundations. RWA, AI, and stablecoins are all accelerating the integration of crypto and traditional finance. RWA bridges the framework and assets, AI solves practical efficiency and execution, and stablecoins are the underlying fuel for all innovations.

✍️ Crypto is transitioning from "single-player games" to "plugins for real-world financial systems." Only those who can connect to real finance will be the long-term winners.

✍️ What truly matters in the AI era is not the models, but the execution:
Payments, settlements, custody, identity, automated strategies.
Whoever provides execution power to agents, whoever's AI agent impresses users will seize the future.

✍️ All short-term surges are expectations; all long-term increases are structures; price is a lagging indicator, while structure is a leading indicator.

✍️ If you cannot understand 2025, you will miss the entire main text of 2026.

❤️❤️❤️ Finally, Jiayi's reflections ❤️❤️❤️

Thank you to 2025, a period that is confusing, perplexing, and filled with self-doubt.
It is this "discomfort" that forces me to pull my gaze away from emotions and re-understand the structural changes happening behind the industry, to see clearly what is true value and what is just noise.

Thanks to the warriors who still choose to BUILD during this cycle.
Your explorations and layouts during the industry's most ambiguous times will be the true starting point of the future "denial of denial"—I believe you will be the ones everyone looks back on in the next five years.

And thanks to myself, and to the team that fights alongside me and endures my daily scrutiny.
Thank you for maintaining curiosity about the industry, a sense of awe towards trends, and uncompromising cognition over the years.

Only do what you believe in; only support those you recognize.
Do not cater to the market's short-term emotions; do not betray long-term structural logic.

The future will still experience repeated fluctuations, but cognition will never decline.