AI still 'has a long way to go'

New technology to integrate deeper and deeper into various industries, say top experts

Development of the artificial intelligence industry still has a long way to go before it can become "real intelligence" and be applied in various sectors, industry experts and market insiders said.

"At the present time, AI is still at the stage of machine learning. It is now driven by data," said Zhang Zhengyou, director of AI Lab & Robotics X at internet conglomerate Tencent.

"A key issue that needs to be addressed for AI development is how to utilize the technology so that machines can know and understand people better," he explained.

He made the remarks at the 2019 Global Artificial Intelligence and Robotics Summit held in Shenzhen from Friday to Sunday.

China is stepping up efforts to develop the AI industry as part of the nation's broader drive to become a global leader in the sector.

The country's technology companies including Tencent, Alibaba and Baidu are competing to gain a foothold in AI as the new technology integrates deeper and deeper into various industries.

"AI is bringing fresh vitality to various industries. It will bring a new technological revolution and unleash enormous energy, pushing forward the world's progress," said Xu Yangsheng, an academician at the Chinese Academy of Engineering.

So far, AI has been used in just a few industries such as finance and security so there is enormous room for industry players to explore, Xu added.

Zhang said Tencent's AI Lab is focusing on applying AI to video games, where the internet giant excels. In addition, the lab is also working on the integration of AI with the agriculture and medical service industries.

"As different industries have different ways to take advantage of AI technology, the application may vary a lot. That's why the current level of the technology is not mature and diversified enough to offer support," said Wen Hao, co-founder of AI startup CloudWalk Technology. "There is still a long way to go before the large-scale application of the technology."

"For example, the financial industry requires a high accuracy rate, while for the security industry, the capability of a quick search is a must. So our major challenge is how to make AI better serve industries and how to make it individualized," he said.

Wen also stressed the importance of privacy in the process, saying that laws and data protection standards have to catch up with rapid technological development.

The Internet Society of China said in a report that there were 3,341 AI enterprises in the nation at the end of last year, accounting for more than 20 percent of the world's total.

The market size of the local AI industry is expected to reach 50 billion yuan ($7.3 billion) by the end of this year and hit 71 billion yuan in 2020, from 33.9 billion yuan last year. The growth represents a 52.8 percent year-on-year rise compared with the same period of an earlier year ago.