Skip to playerSkip to main content
  • 1 day ago
Chinese authorities have held meetings with top tech firms over the past month about potentially restricting overseas access to China's most advanced AI models, including those yet to be released, three people familiar with the discussions said. - REUTERS
Transcript
00:01Chinese authorities have looked into restricting overseas access to the country's top AI models.
00:06That's what sources have told Reuters.
00:09They added authorities held meetings with top tech firms over the past month about the issue.
00:14The talks follow a number of steps by Beijing to keep homegrown AI within the country.
00:19It also shows China, like the US, now treats AI as a critical national asset that needs controls.
00:25The sources said tech giants like Alibaba, ByteDance and ZAI were at the talks.
00:32Chinese AI models have expanded globally since the launch of DeepSeq's R1 model last year.
00:37It's partly due to their low costs and growing capabilities.
00:41Any decision by Beijing to limit access to those products could ripple across AI markets as costs for many businesses
00:47would likely rise.
00:49Two of the sources said the meetings were led by China's Ministry of Commerce
00:52and participants talked about putting limits on the most advanced AI models.
00:57That included both closed source and more open versions.
01:00They added officials talked about making any leak or theft of exclusive AI technology an offence under China's national security
01:07law.
01:08A source further said officials raised the possibility of implementing new measures to restrict who can fund domestic AI startups.
01:15The sources said the scope of the potential restrictions is still being discussed and may only apply to future models.
01:22It wasn't immediately clear when or even if they would come into force.
01:27China's Commerce Ministry and the National Development and Reform Commission didn't respond to Reuters' requests for comment.
01:33Alibaba, ByteDance and ZAI also didn't respond to Reuters' queries.
01:37Three sources also told Reuters DeepSeek is developing its own AI chip.
01:42It's a move which could lower its reliance on Nvidia and Huawei chips.
01:46They said it would be designed for inference rather than for training new models.
01:51Inference is the stage of AI computing where a trained model generates responses for users.
01:56The company didn't respond to a request for comment.
01:59...
Comments

Recommended