Tencent releases AI product for firms as opposition in China heats up

Chinese tech giant Tencent is launching its artificial intelligence product “Hunyuan” for small business use at an annual summit on Thursday, Dowson Tong, CEO of the cloud and smart industries group at Tencent, informed CNBC in an special interview in advance of the celebration.

The news comes days immediately after Baidu exposed a slew of AI-powered apps on Tuesday in the wake of additional supportive regulation.

Tencent has reported it was internally screening its Hunyuan AI product on promoting and fintech. The gaming and social media huge is also set to launch an AI chatbot on Thursday, the enterprise stated in an on line put up.

Tencent is integrating Hunyuan’s capabilities with its existing products and solutions for movie conferencing and social media, Tong informed CNBC.

The company operates WeChat, a widely applied messaging and payments application in China, and video clip convention system Tencent Meeting.

Baidu and quite a few other Chinese providers obtained the eco-friendly gentle in the previous couple months to release AI-run chatbots to the public.

Related to ChatGPT, the bots purport to answer to queries in a human-like, conversational trend — but generally in Chinese. Some, this sort of as Baidu’s Ernie bot, also change text to pictures and movie, with the support of plugins.

OpenAI’s ChatGPT is not formally out there in China. The chatbot releases adhere to new Chinese regulation on generative AI that took influence Aug. 15.

When questioned about the principles, Tong pointed out that this kind of synthetic intelligence is so new that no one is aware of what effects it will have on society.

“It really is prudent to place in some guardrails in location,” he claimed. That will support make absolutely sure the technologies or the companies becoming supplied are of high sufficient high-quality so they really don’t build and distribute untrue facts, he reported.

Chinese authorities reported the “interim” guidelines that took influence final month would not apply to companies acquiring the AI tech as prolonged as the merchandise was not offered to the mass general public.

Which is much more comfortable than a draft introduced in April that mentioned forthcoming regulations would implement even at the investigation stage.

While Beijing has demonstrated it is far more supportive of generative AI than at first feared, Chinese providers also confront U.S. limits on getting superior semiconductors. The most cutting edge versions of the substantial-tech chips, identified as graphics processing units (GPUs), enable companies to prepare AI versions.

“The constraint that we are experiencing will hinder the development, the speed of progress,” Tong instructed CNBC in response to a dilemma about U.S. restrictions.

He pointed out need for computing power overall significantly exceeds supply in China. To mitigate the scarcity, he said companies are “focusing on specific use scenarios, developing products of the correct measurement.”

“And we are hoping that the provide of the GPU compute will be larger in the coming months, and thus the improvement of these technologies can get speedier.”

Tencent is just one particular of several businesses in China — ranging from startups to telephone maker Huawei — that have rushed to announce AI items this calendar year. In August, Alibaba announced it was opening its own AI model to third-bash developers.

Synthetic intelligence involves business-unique education for the technology to generate worth, Tencent’s Tong claimed. He detailed enterprise use conditions in tourism, finance, community expert services and client company.

“We imagine numerous diverse buyers, in truth, would advantage extra by leveraging open up-supply models and use their personal business info to educate for their have styles to meet up with the pretty precise demands in their industrial use conditions,” he mentioned.

That designated use can also assistance with info defense, he said.