Demystifying AI & AI Companies. What foreign policy makers need to know about the global AI industry

Download the Paper (PDF)

As a scientific discipline, Artificial Intelligence (AI) has been around since the late 1950s. During the past few years, however, a subdiscipline of AI known as Machine Learning (ML) has captured the imagination of investors, technologists and policy makers with break through achievements in the recognition of speech and images and its performance in playing complex games. Countries all over the world are adopting national strategies to position themselves to get a piece of the proverbial pie, setting up new public funds to support AI research, expert groups on AI ethics, or new initiatives to study the implications of AI on the future of work.

But AI is not only a hot topic for social and economic policy. It has also become an increasingly important topic in the foreign policy community. For example, as national governments try to strengthen their own AI innovation base, they are worried about domestic AI companies becoming targets of foreign takeovers. But while there has been a lot of talk about AI, the policy debates still suffer from a poor understanding of what AI actually is and means. This paper seeks to address this problem. It explains why most of the international conversation on AI is focused on Machine Learning (ML). In chapter one it breaks down ML into its core inputs – data, software, hardware, and human talent, explaining the importance of each of these inputs for building and applying AI. Chapter two introduces a model with three stages to assess the capabilities of AI companies: early stage ventures, larger companies using AI, and AI platforms.

The analytical framework introduced in this paper can help foreign policy analysts understand AI and develop tools to monitor trends, especially among companies. Foreign policy officers can leverage the distributed network of embassies for information gathering and analysis on the global distribution of ML’s core ingredients and global AI companies. Furthermore, the framework allows foreign policy makers to inform government as a whole on the international AI developments and their economic and ultimately geostrategic implications.

SNV’s Artificial Intelligence and Foreign Policy project was made possible by the generous support from the German Federal Foreign Office and the Mercator Foundation. The views expressed in this policy brief do not necessarily represent the official positions of the German Federal Foreign Office or the Mercator Foundation.