While American leaders fret that China might eventually overtake the U.S. in developing artificial intelligence, Beijing is already way ahead of Washington in enacting rules for the new technology.
Driving the news: Chinese officials will close consultation Wednesday on a second round of generative AI regulation, building on a set of rules governing deepfakes agreed in 2022.
Why it matters: The Biden administration is behind both allies and adversaries on AI guardrails. While officials in Washington talk about delivering user rights and urge CEOs to mitigate risks, Beijing and Brussels are actually delivering rights and mitigating risks.
The big picture: If China can be first on AI governance, it can project those standards and regulations globally, shaping lucrative and pliable markets.
- At the same time, Beijing's speedy regulation achieves three goals at home:
- Delivers tighter central government control of debate.
- Builds up hybrid corporate entities that are meshed with the Chinese Communist Party.
- Boosts trust in AI — already among the highest levels globally — which drives consumer uptake and spurs growth.
Between the lines: After lagging behind the West through consecutive industrial revolutions, leaders in Beijing are driven by a determination not to be humiliated again in the AI era.
- Chinese authorities now have six years of experience building up AI regulatory knowhow since they launched a Next Generation Artificial Intelligence Development Plan in 2017. They're using regulation as a form of industrial policy, in addition to traditional subsidies.
Beijing thinks clear AI rules will help the public trust AI.
- Generative AI's breakthroughs have happened in the U.S., but adoption of consumer-facing AI is widespread in China. Xiaoice, Microsoft’s China-focused chatbot, has 660 million users, and Beijing is betting its faster-paced AI regulation efforts will be a driver for further uptake.
- To minimize social disruption, party leaders accept that the public must support AI and enjoy its benefits.
- "You can't have a vibrant AI ecosystem if consumers don't trust AI products enough to use them," Helen Toner, a director of grants for Georgetown's Center for Security and Emerging Technology, told Axios.
Yes, but: Tech industry insiders see regulation as a drag on speedy AI development, so many view China's lead in this realm as a U.S. advantage.
Be smart: China's AI regulations govern what businesses in China do with AI, but very few in the West believe they will restrain the Chinese government's absolute power in any real way. U.S. efforts to regulate AI take aim at both business and government.
What they're saying: Matt Sheehan, a Chinese AI governance expert at Carnegie Endowment for International Peace, told Axios Beijing's first regulatory efforts were stoked by a fear of AI's downsides, and it has learned a lot over six years.
- "AI regulation is highly iterative, it's about building up the muscles. There's very little chance that you're going to get AI regulation completely right on the first go," Sheehan said.
- After years of tussles, one regulator has emerged to be China's leading AI regulator: the Cyberspace Administration of China, known as CAC.
Details: Authorities have been rolling out rules every few months since 2021, including a national privacy law, the Personal Information Protection Law, and a Code of Ethics for New-generation Artificial Intelligence, which outlines requirements that new AI products deliver everything from "improving human well-being" to "controllability and credibility."
- Since 2022 Chinese AI users have had transparency rights — such as the right to turn off an algorithmic recommendation service, or the right to know when they are being supplied with AI-generated content.
- Beijing rolled out "deep synthesis" rules to protect against deepfakes in 2022. While it didn't anticipate the text-based breakthrough of ChatGPT, those rules provided a head start in drafting new rules addressing the risks of text-based generative AI.
Where's Xi? "Xi Jinping is not waking up in the morning and setting out requirements for generative AI models," Sheehan said. "There's probably no-one on the Politburo in those weeds. They set a high level direction about wanting to be AI leader and then say [to regulators] 'you figure out how to do it'."
The catch: While Beijing has granted rights for Chinese AI users to, for example, challenge an interview turn-down from a hiring algorithm, there's little chance those rights could be exercised at scale, said Russell Wald, policy director at Stanford's center for Human-Centered Artificial Intelligence.
- Chinese corporate and government organizations often don't have the capacity or inclination to deliver on individual rights in this way.
Beijing may also be regulating now and paying later.
- China's path to AI trust comes with hassle and burden for Chinese AI providers, requiring them to filter content and complete security assessments to uphold Communist Party control before new products launch.
- That's the sort of innovation slow-down that U.S. businesses point to as a justification for Washington's relatively more lax regulatory stance.
What's next: Georgetown's Toner expects China's regulations will push its companies to "lean into B2B applications with more limited use cases, since those are easier to train and use within tight guardrails," and China will struggle to "push the frontier of the field forward."
The bottom line: China's top priority is to minimize social disruption as it deploys AI. Beijing has also learned from sanctions on Huawei and U.S. chips export controls that China needs to be ready to innovate alone.
- By moving quickly on regulation, Beijing is creating a foundation for AI exports across the Global South and countries participating in its Belt and Road Initiative.
Flashback: China privacy law leaves U.S. behind
