“Start by introducing AI into your core business, not on the periphery”

As a Founding Fellow of AI Sweden, Göran focuses on international relations, long-term strategy and democratic impact, and brings a significant contribution with his ideas, knowledge and extensive network. His focus areas are leadership and AI transformation. At the moment he is based in Washington DC where he acts as an advisor to private and public sector in North America and Europe that want to develop their businesses in the field of AI. 

What would you suggest to companies that want to take their first steps in the field of AI but haven’t (yet got) started?

I think it’s a good idea to focus on the business aspects in the early stages and not to regard this so much as a technical development. You can begin by looking at the top priorities within your own management team and, on the basis of your requirements, act more quickly than other companies have done. That usually results in a more long-term approach.

It can be good to monitor the situation on a broad basis by focusing on what has been done in your own industry and in other sectors. As AI is a general purpose technology regarding for example pattern matching the finance industry can learn from the healthcare sector and vice versa. You should also look outside Sweden’s borders and see what other examples you can find. I also recommend focusing on core functions within the company at an early stage, rather than choosing areas on the periphery of the business. Often people want to experiment and try things out by having just the real enthusiasts working in the area to be safe. But it’s better to focus on something important because then you need what you’re doing to have an impact right from the start. The company management team must be committed and closely monitor projects rather than completely delegate responsibility.

Which industry is at the cutting edge in AI, apart from the tech sector?

The finance industry got off to a quick start and so did customer service. Legal practices have also done well so far. In the USA, small lawyers’ firms use AI to search through large volumes of data, for example. The retail sector has also made unexpectedly good progress. Another industry in a special position in this field is healthcare. In some areas of healthcare a lot has been achieved and in others hardly anything at all. Many healthcare organizations take an “organic” approach. That is, specialists starts testing AI as one possible tool in their work. This has been the mainstream approach in Sweden but it tends to take really long time to achieve anything. Some healthcare organizations in e.g. USA has taken a more strategic approach to AI. That is, the top management is involved on an almost daily basis and it often results in excellence centers within the organization. The latter is needed to support the different initiatives.

Purely from the perspective of efficiency, which sectors will benefit from speeding up the development of AI?

The sectors that are of interest to me are for example urban planning, construction, local authorities and other similar areas. Here there is a lot of low-hanging fruit and a great deal can be achieved in a short time. In the public sector, including healthcare, a huge amount can be done right now. According to the report produced in January by the Swedish Agency for Digital Government (DIGG), Sweden is wasting around SEK 140 billion every year by not taking advantage of existing AI and by introducing it too slow. The Swedish healthcare sector has an enormous amount to learn from other countries. This is an area where we really should be speeding up the development of AI. Otherwise the answer to the question is really all industries. There is no industry that would not benefit significantly from introducing AI.

How can we resolve the skills mismatch between the companies that will benefit from AI, but know nothing about it, and the AI experts?

I think that the people we describe as AI experts should take a more educational approach. They need to have a better understanding of businesses to be able to introduce the benefits. 

And another approach that I think works even better here in the USA is when business people and economists are brought in who know most about the company. If they learn the business perspective of AI, that is when I have found it simpler to solve the skills mismatch problem. When you can combine this mutual learning you probably get the very best result. 

You are one of our Founding Fellows. What do you think you can bring to AI Sweden and our partners?

I think I can add to the knowledge on how AI is used in different sectors and different parts of the world. I’m also familiar with the political objectives in different countries. More specifically, I’ve dealt with EU issues during the period I spent working in the Swedish Government Offices. I travel a lot in North America and Europe and I have an extensive network. I work closely with the Swedish public sector. I often meet executive teams to support their strategical work on AI. As a keynote speaker on AI I also meet many people which are new to AI.