HomeArticle

The WISE2026 "Annual Figures · Industry Stalwarts" roster is officially released | They march deep into the core of the industry.

未来一氪2026-11-21 20:44
Where the world remains unfinished, in the deep waters of key industries, people leave their respective answers through actions. A person's name thus gains its own unique weight.

 

"In action and speech, people reveal who they are."

Hannah Arendt discussed the uniqueness of human beings in The Human Condition, while the business world is accustomed to defining a generation by prevailing trends. A handful of terms including artificial intelligence, embodied intelligence, and hard technology can easily outline the current landscape of entrepreneurship.

In 2026, these concepts are taking root in specific individuals. Entrepreneurs step onto the stage with their respective technical reserves, seeking opportunities in computing power, chips, robots, and even scientific research tracks with much longer cycles. The business territory is being redivided along the boundaries of technology, and to evaluate a company's potential, it is increasingly necessary to understand exactly what problems the entrepreneurs have chosen to tackle.

But trends cannot explain everything. The same era offers similar opportunities, yet people enter the arena with different knowledge and experience, and the problems they choose gradually become part of their own lives.

This year, the WISE Annual Roster has added a new section "Person of the Year · Industry Trailblazer" to search for pragmatic entrepreneurs who conquer key links in the deep waters of the industry and complete the critical leap of industrialization. Most of the candidates have chosen to solve the most difficult problems in their respective industries, for which there are rarely ready-made solutions to borrow. Entrepreneurs have to make decisions when the answers are still incomplete. Tackling hard problems means taking action to deliver answers at such moments.

An answer often needs to be revised many times, sometimes overturning previous judgments, sometimes restarting with a different approach. And such answers rarely belong to a single person. After all, the answer given by one company often becomes the premise for another company to move forward. To evaluate a trailblazer, we should not only look at what answers they have given, but also at what conditions these answers have created for others.

"Giving answers" is a key thread to understand this roster. Following this thread, you can see where these people set off from and what respective answers they have delivered in different directions.

 

The Decisive Factor Lies Beneath the Model

The intensive emergence of AI can easily create an illusion of uniformity. Dig deeper into the field, some people are trying to make machines understand the world, while others are solving the source of experience required for machine learning. Computing power scheduling, computing-in-memory architecture, embodied data, and robot ontologies correspond to different technical thresholds and customer demands respectively. Some people pursue general capabilities, while others focus most of their energy on the cost or efficiency of a single link. The most popular tracks are accommodating increasingly refined divisions of labor.

Underpinning the division of labor is a hard battle over cost. The National Data Administration disclosed that in March this year, China's daily average token call volume had exceeded 140 trillion, a more than 1000-fold increase in two years. Every call ultimately lands on chips, power supply, and heat dissipation; in 2025, the domestic market share of domestic AI acceleration cards exceeded 40% for the first time.

In the roster, entrepreneurs working on computing power scheduling, computing-in-memory chips, and large model memory are all finding ways to reduce the cost of intelligence at the layer beneath the models: some teams integrate chips from different manufacturers into the same scheduling system, shortening the adaptation of new chips to two working days; some make calculations happen directly where data is stored; others bring their experience in thermal control for satellites into the liquid cooling systems of data centers.

Fields such as overseas marketing and innovative education are also involved in this wave of changes. Some people have incorporated more than ten years of experience in serving overseas enterprises into marketing agents, while others have brought large models into classrooms. When new technologies enter existing businesses, the value of experience needs to be re-recognized. Content can be generated faster, but it is still necessary to understand people in different markets; knowledge is more accessible, but learning still involves the process of comprehension and judgment.

The more refined the division of labor, the higher the requirements for collaboration. The progress of one enterprise may be restricted by unsolved problems of another. Making computing power easier to call, ensuring stable supply of key components, and re-applying experience in new tools — these efforts may not stand in the spotlight, but they are vital to the advancement of the entire industry.

 

Crossing the Threshold to the Real World

Embodied intelligence is the track with the largest number of people in the roster, with nearly half of the shortlisted participants working in this field. This year, humanoid robots have begun large-scale shipments. According to IDC statistics, global shipments of humanoid robots reached about 25,000 units in the first half of 2026, with China's market accounting for 77.9% of the global total; in the same period, the domestic embodied intelligence track received 935 billion yuan in financing, exceeding the total amount of 2025.

After the growth of shipment volume, whether robots can stably complete tasks in different environments still remains to be verified. Coping with the complex changes in the real world requires sufficient training data. The industry generally believes that a general embodied large model requires at least tens of millions of hours of high-quality real interaction data, most of which is deposited in the respective scenarios of enterprises.

Entrepreneurs working on data collection, synthetic data, end-side computing power, cross-ontology large models, and world models are trying to deliver answers from different links respectively. Some teams have launched a million-hour real data program, some let robots learn movements in simulation first before entering battery factories to work around the clock. Others start from scenarios, first figuring out exactly where robots should work. Urban sanitation, charging services, hotel cleaning, and home services are the starting points they have chosen.

The backgrounds of this group of people themselves illustrate the difficulty of the problems. The robotics labs of Huawei and Tencent, research groups from Tsinghua University, Peking University and Zhejiang University, engineering teams from autonomous driving companies... people from different backgrounds have all joined the field. They have already reached high positions in their original fields, but after turning around, they are facing a problem that needs to be solved from scratch.

 

Patience Itself Is a Barrier

Fields such as commercial aerospace and quantum computing extend the time horizon. In the roster, entrepreneurs working on low-orbit satellite constellations, satellite communications, and aerospace thermal control are all in this category; there are also entrepreneurs leading teams to gradually deliver full quantum simulation machines, while continuing to develop general-purpose quantum computers and tackle engineering problems such as continuous operation.

Different tracks have their own rhythms. Some have started responding to specific orders, while others are still verifying their basic capabilities. In the first three quarters of this year, China completed 70 rocket launches, hitting a record high for the same period in history. Amid the popularity, the gaps are also prominent: the total number of satellites planned for the three major low-orbit constellations exceeds 37,000, of which only more than 200 satellites of two constellations have been launched and networked by the end of last year. The situation of "satellites waiting for rockets, rockets waiting for factories" is still the norm in the industry.

The market talks about growth on an annual basis, while technology has its own maturity cycle. Measuring these explorations with the same growth report form will easily make people ignore the unique problems they face. For these entrepreneurs, patience includes the ability to repeatedly revise judgments. Long-term investment does not inherently mean correctness; to keep moving forward, you need more sufficient reasons than the initial enthusiasm.

The hard work in the energy and transportation sectors is also equally long-term. It took perovskite photovoltaics longer than expected to move from the laboratory to the hundred-megawatt production line, and it also took longer than expected for driverless container trucks to drive into port yards. These tracks are the farthest from the spotlight, but once they achieve breakthroughs, they often rewrite the cost curve of the entire industrial chain.

 

Tackling Hard Problems Is Ultimately a Choice Made by People

Age cannot draw a clear boundary for entrepreneurs. There are many post-95s and even post-00s among the shortlisted participants in the roster. Some people started with a self-assembled 3D printer in their university dormitory, embracing the 4D world model; some chose to suspend their doctoral studies and have already achieved commercial applications of dual-arm robots.

From these explorations, we can also see a deeper situation of technology entrepreneurship: The more you try to change the world, the more you have to engage with specific people and matters. One person's knowledge and will will gradually become the work of a group of people, influencing the investment of partners and changing the lives of users.

Technology thus gains social weight, and individuals are placed in a broader network of relationships. A business story may start with one person, but every subsequent step involves the participation of others.

We can see that the shortlisted participants in this roster are already on their way. From the cost of computing power, to whether robots can step out of demo videos and enter factories, to tracks like satellites and quantum that take ten years to verify, every link restricts the forward speed of the entire industry. These problems are difficult not only in technology, but also in time, and returns often appear far in the future.

"Call Me By My Name" — this year's WISE uses this as its theme, trying to bring individuals behind the track labels back into the spotlight. We record their technologies and businesses, and also record a kind of responsibility that cannot be replaced by trends: among many possibilities, choose one path and take responsibility for its consequences. In the unfinished parts of the world, in the deep waters of key industries, people leave their respective answers through actions. A person's name thus gains its unique weight.

At this moment, the 36Kr WISE 2026 King of Commerce Person of the Year · Industry Trailblazer Roster is officially released.

 

This article is from the WeChat official account "36Kr", and is published with authorization from 36Kr.