By DPhil student Konrad Ksiazek
On 22nd November, we had the pleasure of welcoming Professor Anu Bradford as the main speaker for our final Ethics in AI Colloquium this Michaelmas Term, entitled 'The False Choice Between Digital Regulation and Innovation'. Professor Bradford is a Henry L. Moses Professor of Law and International Organizations at Columbia Law School, where she also serves as the Director of the European Legal Studies Center. She is the author of 'The Brussels Effect: How the European Union Rules the World' (OUP 2020) and 'Digital Empires: The Global Battle to Regulate Technology' (OUP 2023).
She was joined by three distinguished commentators. Our first discussant, Dr Caroline Green, is an Early Career Research Fellow at the Institute of Ethics in AI, and a Research Fellow at Reuben College, Oxford. Our second discussant, Professor Ignacio Cofone, is Professor of Law and Regulation of AI based at the Oxford Faculty of Law and the Institute for Ethics in AI. Our final discussant, Professor Carl-Benedikt Frey, is the Dieter Schwarz Associate Professor of AI and Work at the Oxford Internet Institute and a Fellow of Mansfield College, Oxford. He also directs the Future of Work Programme at the Oxford Martin School.
In recent times, we saw a tremendous amount of innovation in the domain of Artificial Intelligence and digital technologies, but it has mostly taken place outside Europe. The need to close the technological gap between the EU and US and China is increasingly seen as an urgent priority. Given this context, the EU's robust regulatory agenda, which was once a source of pride for the bloc, is coming under increased scrutiny and accused of stifling innovation and growth. In her remarks, Professor Bradford challenged this narrative. She rejected the view that we face a binary choice between effective regulation and the commitment to digital rights on one hand, and digital innovation on the other. While she acknowledged the existence of the technological gap in question, she argued that its root causes are different from what is often thought. She noted that there was very little technological regulation in Europe before 2010, at the time when the likes of Meta and Google have been founded. This suggests that regulation was not the major obstacle preventing the rise of similar companies in Europe. She also highlighted that regulation can often promote the right kind of innovation - for instance, the regulation of the AI sphere can improve its market standing by increasing consumer trust with the technology.
Professor Bradford argued that a first major root cause of the technological gap between the EU and its competitors is the absence of a European digital single market. As a result, Europe's digital companies find it harder than their competitors to scale products and increase their customer base. The second root cause of the gap is a lack of a robust and unified capital market in Europe. While European start-ups are relatively successful in securing seed funding, venture capital is far less available on our side of the Atlantic. For example, only 0.01% of Europe's pension funds are invested in venture capital. The third root cause relates to the legal frameworks and cultural attitudes around risk-taking in Europe. Innovation is inherently risky, and European bankruptcy laws are often highly punitive, which discourages bold and speculative investment. This is not the case in the US. The fourth reason behind Europe's technological gap is the EU's relative difficulty with attracting global talent. This is caused, among other things, by attitudes towards immigration and the relative scarcity of world-leading universities in Europe.
In her remarks, Dr Caroline Green observed that regulation can often support innovation by securing the underlying conditions that make it possible. For example, technological innovation cannot happen without patent law and contract law. Secondly, she stressed innovation is not an end in itself - it is only valuable when it improves our lives. Innovation and regulation should go hand in hand to ensure this, and private corporations, investors and ordinary citizens should recognise that they have a stake in ensuring a well-functioning regulatory environment for technological innovation. She reflected on her own research concerning AI-led innovation in the care sector, highlighting the voluntary commitments to various safeguards by some major actors in this area as a positive development.
Professor Ignacio Cofone echoed Professor Bradford's remarks and further challenged the notion that regulation inherently increases the cost of innovation and reduces innovative behaviour. He stressed that much of our private law is indispensable for ensuring the conditions where major corporations can function, and the distinctions often drawn between law and regulation are not always very clear. Furthermore, he observed that there are many examples of AI-specific regulations which are likely to promote innovation and growth. For instance, introducing robust AI explainability requirements can lead to advancements in interpretable machine learning, and regulation can also promote the development of AI-driven bias detection and mitigation systems.
Professor Carl-Benedikt Frey agreed that the European innovation deficit is more long standing than its robust regulatory agenda, and that it is a mistake to rely primarily on the latter to explain the former. However, he stressed that we should not pretend that digital regulation is always costless. He argued that there is strong evidence that the GDPR has increased barriers to entry into Europe, and that stringent labour laws can slow down innovation. He argued that regulation can promote innovation when it also promotes competition, while anti-competitive regulation can stifle it.
This was our final Ethics in AI Colloquium this calendar year. We would like to wish you a restful break, and we hope that you will join our events in 2025.