⇩ Use your ears. Click below to hear this post.
Chinese AI technology provides a blueprint for autocracies around the world.
Chinese artificial-intelligence startup CloudWalk Technology signed a deal in March with the Zimbabwean government, providing the authoritarian regime an advanced facial-recognition system that it can use to identify, track, and monitor citizens. In exchange, CloudWalk gains access to the facial data of the demographically distinct country, which provides the company much-needed data for improving its recognition algorithms.
Arrangements such as this are common under China’s Artificial Intelligence (AI) strategy, whereby Chinese private and state-controlled companies take advantage of the weak legal systems and low privacy standards of developing nations as part of the country’s effort to become a world leader in artificial intelligence by 2030.
Western governments have been rolling out comprehensive AI strategies over the last year, but none fully take into account the geopolitics of AI. The Pentagon is expected to announce its own strategy in the coming weeks and make AI a top priority for the U.S. intelligence community. The U.S. has fallen behind China in a technological race to dominate international applications of AI. Against the illiberal Chinese approach, this new strategy should articulate a global, democratic vision for the technology’s future.
A similar ideological war over technology played out during the Cold War. Concurrently with the American development of ARPANET, the infrastructure that eventually enabled the modern Internet, the Soviet Union developed its All-State Automated System. The lead computer scientist of this competing vision of the Internet was Viktor Glushko, who viewed his invention as a means of resolving the economic-calculation problem, which prevented the efficient allocation of resources from the Politburo and halted the march toward full-throated Communism. Glushko believed that the Russian version of the Internet could resolve their issues with central planning by connecting all people and producing massive amounts of data, leading to indefinite Soviet rule.
The project ultimately failed due to bureaucratic infighting among the Soviet high command, and the West’s free vision of the Internet prevailed. It would be a mistake, however, to believe that a free vision of artificial intelligence will inevitably come to dominate as well.
Against the illiberal Chinese approach, the U.S. must articulate a global, democratic vision for the future of AI.
Research and development in AI is entangled in a global system that allows those with longer-term visions of the technology to shape its implementation. The Department of Defense has struggled to find reliable collaborators in developing its own AI technology, as many of the holding companies that control the necessary intellectual property have opaque chains of ownership often tied to Chinese or Russian interests. The DOD has also had a difficult time finding domestic collaborators to work with them: Google recently pulled out of its contract on Project Maven, which uses AI to more efficiently identify and target terrorists.
While Google’s employees claimed to have ethical problems working on domestic military projects, the company has nonetheless invested heavily in a Beijing research center to attract some of the bright AI talent that China produces. This puts Google in the middle of the Military–Civil Fusionstrategy that the Chinese Communist Party is using to convert consumer technologies to military applications. This complex entanglement of interests may lead Western companies to indirectly fuel Chinese dominance.
AI is an optimization technology, meaning it carries out defined tasks as efficiently as possible. If the goals it is given are set by consumer interests, it has the potential to be one of the most profoundly beneficial technologies ever invented, transforming fields as diverse as transportation, healthcare, education, and energy. If its goals are set, however, by illiberal regimes, it can be used to usher in an era of repression, surveillance, and control.
When Western countries set domestic regulations for technology, this geopolitical consideration should be a priority. Recent developments in the European Union — such as the General Data Protection Regulation (GDPR), which limits the ability of companies to use consumer data for research, and repeated antitrust suits against Google — have pushed investment away from a place where liberal values can play a role in development. That investment instead flows to places where illiberal values dominate. The future of free societies may depend on articulating a vision of AI that reflects liberal values.