Yao Shunyu's First Appearance at the Tencent Conference: Why Join Tencent
```
Author | Huang Yu
On June 5, Tencent’s Chief AI Scientist Yao Shunyu appeared at the 2026 Tencent Cloud AI Industry Application Conference. This was Yao Shunyu’s first appearance and speech at a public Tencent conference since joining the company nearly half a year ago, and the event was packed.
Yao Shunyu’s speech unfolded in a dialogue with Dowson Tong, Senior Executive Vice President of Tencent and CEO of Cloud and Smart Industries Group.
In this conversation, Yao not only shared his views on the “second half” of AI, but also, for the first time, publicly responded to why he joined Tencent.
He said that over the past decades, the logic of AI development has been: for every specific problem, find a specialized solution. For example, for playing Go, invent AlphaGo; for translation, train a translation model.
But now, with pre-training and post-training, the logic has changed as AI moves into its second half.
“It’s as if we now have a universal hammer that can hit any nail,” Yao Shunyu stated. “It’s a general methodology that can solve all kinds of problems. What's more difficult now is finding good problems to solve.”
This is also one of the main reasons Yao chose to join Tencent.
“There are many good problems here, and a lot of products,” Yao said. “This will become increasingly important. What good products can achieve is to generate value by applying pre- and post-trained models, providing context support, etc.”
Apart from the advantage of great products, what persuaded Yao Shunyu to join Tencent was its corporate culture.
“The first time I talked with the board leaders, my first impression was that everyone was very honest. They were very frank about what went well and what didn’t, without any cover-up,” he said. “Overall, Tencent is a company operated based on trust, not metrics.”
This is extremely important for AI, because AI isn’t a business that can be simply driven by KPIs.
Yao Shunyu also set a clear personal goal: to build a long-term AGI-based organization in China.
The Hy3 preview (Hunyuan 3.0 preview), released in April, was seen as Yao’s first report card since joining Tencent.
On this, Yao Shunyu commented: “There is no secret. To be frank, building large models today is, to some extent, a trivial task; the key is to do a good job on infrastructure and data—the algorithm part is actually relatively simple.”
At the conference, Dowson Tong didn’t shy away from raising the question of external doubts about “Tencent lagging behind in AI.”
Yao Shunyu also shared his view.
Yao pointed out: “In the past, we explored many models and products, took many detours. I think this is normal—if you haven’t done something before, the first time is bound to have twists and turns. But what’s more important is whether you can be honest with yourself, see the feedback and change, and remain patient. This is the most important thing in the second half.”
Yao believes AI is a long-term game and the second half has only just begun. He doesn’t think ChatGPT and CloudCode will be the sole super apps—there will definitely be a continuous stream of new opportunities. “Maybe today is just like when PCs first appeared in the 1970s; I think there is still so much more to do.”
Another observation is that he thinks this is a pluralistic game—undoubtedly, Coding Agents’ productivity will become more important, but there is still much room left to fill.
Risk disclaimer and liability clauseThe market carries risks; investment needs caution. This article does not constitute personal investment advice and does not take into account individual users’ specific investment targets, financial situations, or needs. Users should consider whether any opinions, views, or conclusions in this article are suitable for their situation. Investing based on this is at your own risk. ```