Why did I gradually give up on researching Web3 projects?

Introduction

Recently, I have shifted most of my focus to the AI field, resulting in a decrease in my output related to Web3. However, after more than a year of reflection, I have accumulated many new insights and experiences about this industry that are worth sharing.

Readers who have followed me for a while may remember that my writing career began with research and analysis of projects and sectors. But I can’t recall when I started to write fewer articles of this kind. Behind this change lies both an enhancement of my personal perspective—allowing me to glimpse the higher-level and more fundamental operational logic of the Web3 world—and a series of shifts in my personal resources and wealth concepts.

During this time, I have been frequently asked by friends: "How is this project?" "Is that sector still worth investing in?" I often find myself at a loss for words because, in the current environment, these questions have become difficult to answer definitively.

After some time of thinking and sorting through my thoughts, I would like to systematically discuss why my enthusiasm for research and analysis of specific projects has gradually shifted towards abandonment.

Core Idea 1: The Reversal of Information Barriers—When AI Becomes a Tool for Creating Fog

It is undeniable that a core profit model in the Web3 industry stems from information asymmetry. In terms of "research and analysis," those who can discover a project's potential value earlier and position themselves accordingly can reap excess returns. However, it is precisely this reason that led me to ultimately abandon this path.

Looking back to 2018 and 2019, I was still engaged in project ratings. Thanks to my background in computer science, many blockchain concepts that seemed obscure to outsiders were quite familiar to me. This allowed me to relatively easily distinguish which projects were hollow and which truly possessed technological value.

However, as we reach 2025 (referring to the current and near-future industry environment), this methodology has nearly become obsolete. It is not that the development of blockchain technology has surpassed my understanding, but rather that project teams have become incredibly adept at using the latest AI large models to "package" themselves. Projects that could once be easily seen through have, with the aid of AI, crafted narratives, technical white papers, and even GitHub repositories that are seamless and convincing.

I can candidly share that over the past two years, I have helped some exchanges and project teams write many promotional materials that appear "technically professional" to the outside world, but their true authors were actually AI. Even the seemingly active project interaction data and on-chain transaction records were often generated through scripts written by AI.

This means that in the era of AI proliferation, the cost of traditional research and analysis is increasing exponentially. To discern the authenticity of a project, the effort and time required far exceed what was needed in the past. Public information channels have been severely polluted by AI-generated "noise," and it feels as if we are watching a "magic duel" between AIs, while real and effective information is obscured layer by layer. I have personally attempted to use AI to analyze Web3 projects, but progress has been minimal, and I feel trapped in a feedback loop of AI-generated content validating each other.

Core Idea 2: Decoupling of Value—The Disconnection Between Project Quality and Token Price

For many who have not deeply engaged in Web3 research and analysis, this seems like a high-return path. Indeed, in the first two cycles, I earned considerable profits through research and analysis. But that was in a relatively "pure" era of the industry—good projects truly appreciated.

Today, Web3 has evolved into a highly mature and clearly divided industrial chain. From project preparation, fundraising, issuance, promotion to market value management, every link has professional institutions or incubators operating behind the scenes. Even many KOLs you see have the support of exchanges.

As an independent researcher "on the outside," the possibility of conducting research and profiting solely based on public information has become minuscule.

A deeper issue is that in the vast majority of Web3 projects, the technical team and the operational team are separate. In other words, there may indeed be a group of tech geeks dedicated to building excellent technology, but the price trends of the tokens are not determined by them. During the fundraising phase of a project, the market-making rights of the tokens are often handed over to professional operational teams.

Therefore, when a project announces significant good news, such as a technological breakthrough, it may actually be an excellent opportunity for the operational team to distribute tokens. This explains the frequently seen phenomenon: why does the price drop when technology has broken through?

Ultimately, the industry has evolved into its current state: the quality of the project itself and its token price performance are completely unrelated. This is the fundamental reason why I find myself at a loss when friends ask me questions like "Is the project good? Can I buy the token?"

Core Idea 3: The Disappearance of Fundamentals—An Era Where Traffic and Emotion Reign Supreme

This point may be the most painful: in today's era of rampant meme culture, the quality of the project itself has become unimportant. Project teams do not care, and most participants do not care either. Traffic and emotion have become the only metrics for measuring a project's success.

I myself am also following some projects, such as the highly anticipated Monad ecosystem that is about to airdrop, but its overall heat and community participation may far pale in comparison to some suddenly popular meme projects.

This precisely reveals a cruel characteristic of Web3 today: "I came to Web3 to make money; my goal is profit, not to build a quality project." When the consensus of the entire market is built on this, in-depth research into the project's fundamentals becomes trivial, even somewhat "out of place."

On the other hand, as I engage with higher levels of the industry, I gradually realize that many project teams do not consider the quality of the project a key topic when negotiating with investors or operational institutions. As long as they choose a sector that sounds good and has heat, and weave a compelling narrative with AI, the rest is merely a game of social dynamics and token distribution. As for the project's development progress, that is merely a timeline they use to decide when to distribute tokens.

Conclusion: The True Value of Research and Analysis

The purpose of writing this article is not to completely deny the value of "research and analysis." On the contrary, research and analysis play an immeasurable role in broadening personal perspectives, enhancing cognitive depth, and building knowledge systems. It has at least allowed me to grow from a naive "retail investor" into a participant who can avoid most traps.

However, if your sole purpose is short-term profit, then I believe that in the current era, relying solely on public information for research and analysis to make money has become an exceptionally narrow path.

Today, the nature of publicly available research and analysis content has evolved more into a "traffic tool." For example, I once spent a month operating a research account, and the articles easily reached tens of thousands of views. But the endpoint of this path often leads to driving traffic to third-party paid communities, which then guide you to purchase certain tokens through various means, with the ultimate profit point still resting on "selling tokens." Because I believe this model is not honorable and did not yield profits for me, I later abandoned it.

My experiences in research and analysis over the years have given me an unprecedented understanding of Buffett's famous saying:

"Never invest in a business you cannot understand."

In the past, I thought "understanding" meant comprehending the technology and model. Now I realize that in Web3, "understanding" must also include grasping the underlying capital structure, interest games, and human nature. And these are precisely what public information can never reveal to you.