The Secretary General’s blog
A technological revolution with fantastic opportunities is just around the corner. The future will see us surrounded by artificial intelligence, which has been described as the new electricity and will be found in virtually all machines and industries. Artificial intelligence (AI) is capable of so much – providing tailor-made recommendations on the internet, helping doctors treat patients and driving autonomous vehicles, to name but a few. But how should we address the ethical dilemmas associated with it? How should we go about maintaining our much coveted high levels of social trust in the Nordic Region?
A rather brief definition of AI is that machines appear to be intelligent as they attempt to imitate human thinking and behaviour in order to get something done. Computer programs capable of learning from large volumes of data and their own experiences have attracted a lot of attention recently.
AI has so much to offer. It can be used to save lives in the health service, combat climate change, make improvements in the public sector and to trigger growth for Nordic business. We need to be part of this. We need to seize the opportunities. If the Nordic Region is to make progress and not put our high levels of social trust at risk, it is crucial that we deal in an open and honest manner with the ethical dilemmas and potential clashes with our values that large-scale use of AI might entail.
In 2017, the Nordic Council of Ministers for Digitalisation announced that it wants the Nordic and Baltic region to be a digital leader and show the way in Europe. The ministers also agreed to work together to develop and promote the use of AI in the Nordic Region, e.g. via skills enhancement and better access to data.
One of the reasons it will be important to discuss AI and ethics from a Nordic perspective is because of the potential consequences for our Nordic values. In response to a questionnaire survey conducted by the Nordic Council of Ministers and the Nordic Council in 2017, people in the Region identified freedom of speech, openness, democracy, shared values and equal rights as typical Nordic values.
The ministers have taken these messages to heart, and the Nordic Council of Ministers for Digitalisation is drawing up ethical guidelines, standards, principles and values for when and how AI should be deployed. These Nordic conclusions will be fed into the EU Commission, which is also working on ethical guidelines for the development of new technology.
As part of this process, it is important to think about how we will maintain our democratic values. One concern is that AI can be used to manipulate elections by deploying subtle mechanisms based on machine learning and big data. On the other hand, there is the prospect of AI improving democracy; for example, it can also be used to combat fake news.
As AI becomes more common, it may also affect openness and transparency in the Nordic countries. The most powerful forms of AI are often referred to as a black box because it is difficult or impossible to work out how the system came up with its solution. We need to ask ourselves whether AI will weaken trust in officialdom if we no longer know how the technology with which we are surrounded actually works and if it makes decisions that we do not understand.
We also have to think about the consequences for the value of the individual, for the value of human dignity. How will we value the human ability to think and to reason when machines take over more and more jobs and do them faster and smarter than we do and in ways that we do not understand? What value will human beings, and the human mind, have in a world in which machines are smarter and faster at most things? And, if AI assumes responsibility for more and more roles previously the domain of humans, will this make us look at ourselves and others differently? How will AI change the place of human beings in society?
We also have to take into account the fact that AI may discriminate. Algorithms are capable of perpetuating historical prejudices found in the data they use. In other words, the opinions and values in the data from which AI learns may clash with Nordic values. How will we avoid this problem?
Another issue that we have to face up to is ensuring that AI is used to promote creativity and innovative thinking rather than make us more passive? The fact that AI will simplify lives also means that there is a risk of it having a detrimental effect on our cognitive functions. As machines assume responsibility for more and more tasks, there will be less need for people to use their problem-solving abilities. We run the risk of our senses becoming less responsive, our ability to pay attention and understand being reduced and of us becoming more dependent on technology.
On what basis upon should machines take decisions about us? One oft-repeated but illustrative example is the autonomous vehicle that has to choose between two alternative courses of action, both of which will lead to the loss of human life. What will it choose? What values will be programmed into it and by whom? We also have to compare this unpleasant scenario with the fact that autonomous vehicles will reduce the number of traffic accidents quite considerably and save large numbers of human lives because they will not make the same mistakes that humans do.
All of these ethical problems demonstrate the need to think consciously about technological developments, a point not lost on our Nordic ministers. It is not a matter of putting on the brakes but of finding the right balance. We must pave the way for AI to generate growth and improve lives in the Nordic Region but also protect our much coveted social trust and our respect for the fact that all people are of equal value. The challenges are great, but so are the opportunities.