China’s AI strategy focuses on economic transformation while addressing concrete risks, an approach with which Hong Kong, in its five-year plan, is aligned
apocalyptic warningsfrom Silicon Valley’s leading AI bosses. While American executives speak of slowing development, the city and mainland China are systematically embedding AI across industries – a measured strategy focused on tangible economic transformation.
China tightly regulates AI and so largely avoids the existential worries the technology provokes in the US. Recently, Anthropic CEO Dario Amodei said that “we must slow the pace at which we improve the capabilities of AI models”; rival OpenAI’s Sam Altman agreed. By effectively dividing AI development into “responsible” systems regulated by trusted guardians versus “dangerous” open-source models requiring constraints, they establish a binary that advantages their business models. Calls to slow AI’s development squarely target China’s progress, especially since Amodei is well known as an AI hawk. Beijing agrees to the need for global AI governance through a multilateral organisation like the United Nations. It objects to any nation or tech insider unilaterally dictating rules and imposing them on others.
The safety discourse serves domestic American purposes. If catastrophic AI risk becomes the dominant narrative, who regulates it? Not government bureaucrats lacking technical knowledge, but the very companies issuing warnings, the same actors who create the alleged threats in the first place. This allows frontier labs to position themselves as indispensable gatekeepers while gaining autonomy from oversight.
Legitimate safety concerns exist, and China takes them very seriously. The distinction lies in approach.