1. What is the competitive landscape?
It depends on what you measure. The US has largely taken the lead in the generative AI space, with its biggest public breakthrough coming in November when San Francisco-based OpenAI launched his ChatGPT chatbot. This has sparked a wave of investment in Silicon Valley, with Alphabet’s Google and other rivals competing with OpenAI and its backer Microsoft to develop commercial applications using more sophisticated AI tools. ing. China is seen to be more advanced in areas such as image recognition, at least when it comes to commercialization. So far, most demonstrations of China’s generative AI tools suggest it has a long way to go. But the last few months have seen an explosion of startup creation. Virtually every major Chinese tech company, from Baidu Inc. and ByteDance Ltd. to Tencent Holdings Ltd. and Alibaba Group Holding Ltd., has announced his ChatGPT candidates or announced plans to do so. In late June, Baidu announced that its ChatGPT-style service outperformed OpenAI’s out-of-the-box offering in several ways.
2. What do the numbers represent?
The U.S. led the way in AI investment, attracting $26.6 billion in the year to mid-June, compared with $4 billion in China, according to consulting firm Preqin. The gap is slowly closing, at least when it comes to transaction flow. There were about 447 AI venture deals in the US during the same period, but China accounted for more than two-thirds of all US deals, up from about half in the past two years. From a research perspective, China has been the leader in AI journal citations since 2010. According to Stanford University’s latest Artificial Intelligence Index report, China’s citation share in AI publications will be nearly 30% by 2021, almost double that of the United States. . China has historically lagged behind when it comes to turning ideas into real-world entrepreneurship. The report found that between 2013 and 2022, the United States had the highest number of newly funded AI companies of any region, with more than 4,600, 3.5 times more than China during the same period.
3. Why are China’s AI advances worrying US officials?
China’s watchdogs have long used AI to crack down on crime and social unrest, and are said to help the country’s security agencies monitor the Uyghur minority in the Xinjiang Uighur Autonomous Region. As a result, the US has imposed sanctions on Chinese companies such as SenseTime Group and Hangzhou Hikvision Digital Technology. China’s military use of AI is also a cause for concern. In October, the Biden administration introduced a series of restrictions restricting U.S. chip exports to China over concerns that China could use advanced semiconductors and AI to improve its military capabilities. In November, the Pentagon released a report on China’s military and security developments, detailing Beijing’s vision of “intelligent warfare” defined by the increased use of AI and other technologies at all levels of combat. described.
4. What is the US doing about it?
The U.S. government’s approach has gradually shifted from targeted sanctions aimed at curbing the development of Chinese products that could threaten U.S. national security, to broader efforts to protect U.S. technological superiority. are moving to Longer term, the United States is considering whether to impose investment restrictions on Chinese tech companies. This will have significant ramifications, as China has long relied on US money to fund its most promising start-ups. Chinese technology is seen as a goldmine for U.S. investors, with a focus on commercializing AI and applying it to existing business models. This contrasts with the approach of companies like Google, which has spent years researching AI in depth, open sourcing some of its findings, and continuing to focus on removing barriers to innovation.
5. What are China’s limits in AI?
First, there is the issue of access to US hardware. Powerful chipsets from Nvidia Corp. and Advanced Micro Devices Inc. are essential for training AI models at scale, he said, but Washington state has banned these most powerful chipsets from China. I’m here. Google Research scientist Zhou Yanqi said at an AI industry conference in June that a U.S. chip ban would be a “huge detriment to Chinese startups.” Chinese tech executives argue that their generative AI models can still perform well if they combine a number of underperforming chipsets to compensate for the decline in performance. China has a wealth of hardware needed to tackle AI. Billionaire Baidu founder Robin Li said the United States and China together account for about a third of the world’s computing power. Skeptics of China’s technological ambitions argue that true innovation requires the unbridled exploration and experimentation cultivated in the United States, but suppressed in China. Extensive censorship, they argue, means that the datasets used by Chinese applicants are inherently flawed and artificially limited. But past experience shows that China’s relative isolation is no barrier to technological success. Tencent’s WeChat became a social media giant in the absence of Facebook and Twitter, while Baidu became China’s leader in domestic online search after Google’s exit.
6. What about regulations?
Given the risks of AI platforms becoming overwhelmed with disinformation or becoming tools for mass violations of privacy and content rights, one of the wildcards in this race is national approaches to regulation. China moved first to regulate the AI industry, which is not surprising given the government’s preoccupation with controlling online content. It will require a security review of AI services before allowing them to operate. The draft regulation suggests that the responsibility for training AI algorithms and enforcing censorship will fall on platform providers, a significant burden. The United States has begun to move toward regulation. But a parliamentary deadlock is making it difficult to enact meaningful AI policy.
More articles like this can be found at bloomberg.com
