How to invest in potential stocks of the AI era?

Recently, a tech blogger I often follow shared this perspective in a private course:

If the invention of ChatGPT is the "steam engine," then the invention of OpenClaw is the "crankshaft."

This blogger frequently uses the examples of the steam engine and crankshaft when discussing the role of technological progress in human society.

Although the invention of the steam engine was a significant breakthrough in the history of human technology, its use in human society did not become widespread and did not have a substantial driving and promoting effect in the one or two decades following its invention. It was another overlooked invention—the crankshaft—that truly enabled the steam engine to achieve large-scale popularization and application.

After the crankshaft was invented, human society began to use the steam engine in automobiles, ships, and trains, in various large industries, and ultimately, the steam engine began to genuinely drive the progress and development of human society.

In the same vein, the invention of ChatGPT is an important breakthrough in the history of AI development, but its invention seems to have only been used and circulated in a limited scope; the general public has not felt the presence of AI nor the direct changes it brings. However, OpenClaw is likely to become a turning point in the field of AI, facilitating large-scale applications of AI.

There is an important distinction between OpenClaw and previous AI Agents:

AI Agents require interaction with humans at every step, and they will stop when they encounter a problem.

But OpenClaw is different; when it encounters a problem, it can continuously try new methods through self-backtracking. This makes it possible for AI to complete tasks on its own without human intervention, removing a key obstacle for large-scale applications of AI.

Of course, OpenClaw still has quite a few issues and requires a lot of refinement and evolution, but this trend is already very evident.

Furthermore, this blogger believes that as early as this year, with the promotion of OpenClaw's application, there is likely to be disruptive applications aimed at a broad user base in the field of AI, so we need to pay close attention to its development and subsequent changes.

After listening to this sharing, I thought of another question:

What kind of companies will develop truly disruptive applications aimed at a broad user base?

This is a question I have been pondering.

Why am I so focused on this question?

Because it is closely related to investment; it determines where to focus attention when analyzing and observing various companies.

Since people's time, energy, and resources are always limited, we cannot spread our attention across all companies. Trying to grasp everything will likely result in grasping nothing.

In the past, when I shared the development history of blockchain, mobile internet, and internet ecosystems, I mentioned that after the emergence of each new ecosystem, the vast majority of the ultimate giants in the new ecosystem are newly emerging companies, and very few old players from past ecosystems can continue to maintain their position in the top tier of the new ecosystem.

There are very few companies like Tencent that span both the internet and mobile internet and continue to dominate.

I feel this pattern may continue to apply in the AI era.

In the AI era, although past giants (such as Microsoft, Google, Meta) are still frantically pouring money into this field to seize territory, it currently appears that emerging giants are leading the trend.

In the large model field, OpenAI and Anthropic are typical examples. In our country, DeepSeek has pioneered the way, and the most notable companies in the Hong Kong stock market, such as Zhiyu and MiniMax, are also examples.

Although Google's Gemini and Alibaba's Qianwen are now competing with those giants and ranking high on various large model leaderboards, I still feel that the ultimate top winners will likely belong to the aforementioned companies.

Let’s specifically mention Google's Gemini. Its core technology actually comes from its acquired company DeepMind, rather than from itself. So strictly speaking, Gemini is also a product of a new company.

In the application field, the recently popular Manus and the newly emerging OpenClaw and Moltbook all come from new companies.

If OpenClaw is indeed the "crankshaft" of the AI era, then in the upcoming transformation, we are about to see the "automobiles," "ships," and "trains" of the AI era—created by a group of currently obscure small teams and new companies.