Beijing is set to enforce new regulations for artificial intelligence (AI) services this week, with the aim of striking a balance between state control and global competitiveness. The Chinese government has issued 24 guidelines that require platform providers to register their services and undergo security reviews before launching. Oversight will be carried out by seven agencies, including the Cyberspace Administration of China and the National Development and Reform Commission. While the regulations are less onerous than the original draft from April, they reflect China’s commitment to government oversight of AI, following the lead of Europe. In contrast, the US has no significant legislation under consideration, despite industry warnings about the risks of AI. China’s regulations go beyond those contemplated in Western democracies, but include elements that enjoy support in the US. For instance, Beijing will require conspicuous labels on artificially created content, such as photos and videos, to prevent deception. Chinese companies will also be required to use legitimate data to train AI models and disclose that data to regulators as necessary.
The US has proposed a range of AI regulations, but these efforts are in their early stages. The US Senate has held AI briefings this summer to familiarize members with the technology and its risks before pursuing regulations. In June, the European Parliament passed a draft of the AI Act, which would introduce new guardrails and transparency requirements for AI systems. The final terms of the legislation will be negotiated between the parliament, EU member states, and the European Commission. China has been laying the groundwork for these regulations since 2017, when the State Council declared the development of AI a priority. The government has since consulted with legal scholars and undertaken months of consultation with regulators, industry players, and academics to strike a balance between legislation and innovation.
China sees AI as one of its tech priorities and wants to use it to boost its economy and compete with the US. The country has enlisted the help of major players such as Baidu, Alibaba, and SenseTime, all of which have showcased their AI models. However, Chinese companies still lag behind global leaders like OpenAI and Google. The US has opted for a more hands-off approach to regulation, allowing Silicon Valley giants to thrive, but this strategy is seen as dangerous with generative AI. The lack of oversight may lead to risks and threats to humanity, according to experts. China’s regulations aim to strike a balance between supporting their AI ecosystem and maintaining social control and the ability to regulate and censor information.
The different approaches taken by the US and China are reflected in the way companies handle information. OpenAI’s ChatGPT, for example, has been accused of generating false information and has raised concerns about its lack of control. Chinese companies, on the other hand, must be more careful due to the strict censorship regime in the country. However, technical challenges remain as AI innovations can often go beyond the control of platform service providers. In China, some companies are using two-way keyword filtering to scrub controversial content. The government also conducts spot-checks on AI services’ labeling of data. The European Union has taken a progressive stance on protecting individuals from AI overreach, with a draft law that emphasizes privacy controls and restricts the use of facial recognition software.
Credit: The Star : Tech Feed
