Debating the future of artificial intelligence

Reported by Rebecca Van Hove, Policy Fellowships Administrator

Throughout the nineteenth and twentieth centuries, technological changes and advancements in machine innovation have habitually been celebrated as signs of human intelligence and human progress.

In the twenty-first century, however, this is no longer automatically the case: machine innovation and advancements in artificial intelligence may now have the potential to be decoupled from human intelligence itself - the rise of the former even potentially posing a threat to the latter.

On 5 July 2018, CSaP collaborated with the Mercator Science Policy Fellowship Programme to host a public debate in Brussels on the future of artificial intelligence (AI).

Held at the Representation of the State of Hessen to the European Union, the event brought together experts on AI from Cambridge and the Rhine-Main universities for a debate on how artificial intelligence will transform public services and business and what role policy makers can play in preparing society for the future impact of artificial intelligence.

The event was chaired by Robert Madelin, chairman of consultancy network Fipra International, who has experience of policy-making on emerging technologies as former Director General of the Directorate-General for Communications Networks, Content and Technology at the European Commission. Panellists included: Professor Iryna Gurevych, Director of the Ubiquitous Knowledge Processing Lab, Technische Universität Darmstadt; Professor Oliver Hinz, Chair of Information Systems and Information Management, Goethe-Universität Frankfurt; Dr Julian Huppert, Director, Intellectual Forum, Jesus College, Cambridge; and Professor Thomas Metzinger, Director of Theoretical Philosophy, Johannes Gutenberg-Universität Mainz.

AI will have significant implications for the world of work. Iryna Gurevych argued that artificial intelligence would have a role to play in all professions in the future, also emphasising that AI is more than only machine learning through big data. She argued for the need to invest in schooling to prepare people for the societal changes being wrought by technological changes.

Oliver Hinz similarly pressed for the importance of improving university education and providing relevant skill training. There is an urgent need to understand the changes currently taking place in so many aspects of our daily lives, and thereby also in the way in which government, services, companies and communication function. Building students’ knowledge and skills in programming and data analysis is crucial in order to counter a growing skills gap.

Julian Huppert continued the discussion of AI’s impact on future work from a different perspective, arguing that these developments could also offer an opportunity to rethink the future of humanity in a positive manner. Currently, Julian contended, we are heavily defined by our jobs - but does it have to be like that? Rather than starting out from the wrong place, with a focus on employment as a goal in and of itself, Julian questioned whether we actually need to generate enough jobs for everyone. Instead, could AI allow us to lead a different type of life, in which work is perhaps no longer as central?

Yet alongside this more optimistic note, Julian also stressed the possibilities of negative consequences of AI. He argued that – despite its unlikeliness – it is necessary to seriously consider the possibility of human destruction or even extinction, simply because the stakes are so high and potentially catastrophic.

Thomas Metzinger made the point that artificial intelligence brings with it very different predictive horizons and that it is important to separate these out in our debates on how to tackle such potentialities. The less certain possibility that a super-human intelligence will exist by 2070, for example, is very different from the closer and likelier possibility that AI will contribute to a huge spike in unemployment in 2030. These are both issues which governments and policy makers should deal with, but in different ways.

Thomas also argued for an ethical European-wide ban on all artificial consciousness research, as well as a ban on all autonomous weapon systems. Oliver Hinz contended that the EU should prioritise funding research on the use of AI in medicine, for example in early diagnosis. As the use of personal data for medical research carries a large consensus of general support, he believed the application of AI in medical care should be further encouraged.

Thomas Metzinger also advocated for timely action from political institutions to ensure that AI does not enhance the societal income gap. The panellists agreed that the current landscape of big data analysis sees too much power concentrated in industry – as big data becomes more and more important as a strategic tool, they see a growing need for regulation of the collection and use of data. Thomas contended that rather than becoming competitor to other countries in terms of funding AI research, Europe should keep its focus on the ethical and legal dimensions of artificial intelligence: Europe’s unique selling point, he argued, are precisely its ethics and its law.

Questions from the lively audience ranged from whether we should tax AI and robots, and the effect of AI on societal equality, to the need for regulation of big data and how to deal with fake news. Robert Madelin concluded the discussion by highlighting that any discussion of the future of AI really is a discussion of what we as humans want our own future – to be set in a world of AI – to be. As such, as the panellists argued, we will need to focus on education and skills building in order to better prepare for an AI future, while also crucially paying enough attention to the ethical dimensions of new technological developments.

About the Mercator Science Policy Fellowship

The Mercator Science Policy Fellowship is a demand-driven and tailor-made programme for senior policy professionals and journalists in Germany, hosted by the Rhine-Main universities of Frankfurt, Mainz and Darmstadt. It was set up in 2016 inspired by CSaP’s Policy Fellowship in Cambridge and is funded by the German Mercator Foundation.