AI’s Potential Is Being Curbed by Lack of Trust and Talent, Experts Say
By Wang Xintong
The artificial intelligence (AI) industry, still in its early stages, holds significant investment potential and trillion-dollar opportunities in fields like humanoid robots and AI agents. However, its adoption is hampered by challenges such as data integration, a lack of trust and a talent shortage, according to speakers at the Asia New Vision Forum 2025.
During a panel discussion Thursday at the three-day forum organized by Caixin Global, Xiong Youjun, general manager of the Beijing Innovation Center of Humanoid Robotics, said that despite a surge in interest, the humanoid robotics sector is far from overheating, especially in China.
Total investment in China’s robotics startups “may have just reached tens of billions of yuan,” compared with the billions of dollars raised annually by some North American counterparts, he said.
“We believe that humanoid robots are still at a very early stage,” Xiong said, describing today’s investments as just an appetizer, merely a taste of what’s to come.
With many companies having begun small-scale market entries this year, mass production is expected to begin next year, he said, adding that this will require far greater funding in areas like core components and large language models (LLMs).
“We believe the bulk of investment is still ahead of us. The day humanoid robots truly enter households, this market will likely be valued not in the tens or hundreds of billions today, but in trillions of U.S. dollars,” he noted.
Zhao Yuli, chief strategy officer at Beijing-based robotics startup Galbot, echoed this optimism, asking, “who wouldn’t want a robot at home to act as their housekeeper?” She cited a recent Morgan Stanley report projecting China’s humanoid robot market will grow to $40 billion by 2035, up from just 1% of that size in 2024.
“(The industry’s) growth momentum is incredibly strong.” she said. “The real challenge isn’t a lack of demand — it’s about who can effectively connect supply with demand. That comes down to a company’s ability to transform technology into viable products and ultimately scale those products to meet real industrial needs. That’s what will determine who succeeds in the long run.”
AI agents
At another panel discussion Thursday, Cynthia Zhang, founder of FutureX Capital, which focuses on investing in China’s growing high-tech firms, said, “AI is fully ready, but we don’t have a lot of great products yet.”
This gap is particularly evident in the development of AI agents.
Since early 2024, tech giants around the world — from Google LLC, Microsoft Corp. and Anthropic PBC to Baidu Inc., Lenovo Group Ltd. and Zhipu AI — have raced to introduce their own AI agents. Once marketed as enterprise productivity tools for coding and website development, agents are now being targeted at consumers, helping users open apps, complete multi-step tasks and manage digital life in a more intuitive, less robotic way.
In March this year, the general-purpose AI agent Manus, developed by Beijing Butterfly Effect Technology Co. Ltd., captured global attention. Hailed as the first truly autonomous AI agent to enter the mainstream, it can independently perform complex tasks such as filtering resumes and analyzing stocks.
Zhang’s firm aims to invest in 100 AI agent teams but has so far only backed 12, including the team behind Manus.
She noted that the industry lacks engineering teams with deep technological understanding, agility and data access. Zhang also highlighted the difficulties startups face in obtaining high-quality, domain-specific data and in gaining the trust of enterprises for co-developing products.
However, she believes this won’t be a bottleneck for long: “we see a lot of opportunities in the next 5 to 10 years.”
Investors are placing bets across the entire AI stack, from foundational models to applications. Nevertheless, the quality of the founders is often considered a more critical factor for investment than the specific segment of the industry.
FutureX Capital has invested in 60 AI companies across China and the U.S. over the past two and a half years, with 70% of them already profitable. “It’s not about which sector you bet on, it’s not about when to go, when to invest, it’s that you should go all in … but most importantly, you really need to bet on founders,” Zhang said.
Hans Dekkers, general manager of IBM Asia Pacific, also expressed confidence in the global AI industry. He believes there is still room for multiple LLMs, as well as thousands of domain-specific smaller models trained on proprietary enterprise data.
Challenges
Still, the path to widespread AI adoption remains fraught with challenges, particularly around trust, safety and data integration, according to speakers at the two panels.
On one hand, people need to trust the reliability of AI technology, much like how they gradually come to trust autonomous driving systems, said Aaron Tan, CEO of Southeast Asian online car marketplace Carro. On the other hand, it is crucial to alleviate fears around AI by helping employees understand that its purpose is not to replace jobs, but rather to boost work efficiency, he said.
Wang Haojun, chief financial officer of Chinese autonomous driving firm Pony.ai Inc., emphasized that in autonomous driving, safety isn’t just a technical issue but a public and regulatory one. “Without real-world data demonstrating that autonomous driving is significantly safer than human driving, its adoption will face resistance from both the public and regulators.”
Many large organizations remain siloed, using different software products and creating their own AI agents, said Dekkers. He sees a huge opportunity in “integrating all these agents into one single platform.”
“AI is just at the beginning of exploding,” Zhang said, adding that every industry will be reshaped by AI, it’s just a matter of “which company you choose to reshape the industry.”
Contact reporter Wang Xintong (xintongwang@caixin.com) and editor Jonathan Breen (jonathanbreen@caixin.com)
caixinglobal.com is the English-language online news portal of Chinese financial and business news media group Caixin. Global Neighbours is authorized to reprint this article.
Image: Johannes – stock.adobe.com