Towards trustworthy AI and an autonomous Europe

Professor Michela Milano is the Deputy President of EurAI and one the keynote speakers at AI Day 2020
Europe and EU's advantages include the strong university system, the long history in conducting research of top quality, and an industrial ecosystem that includes start-ups, big corporations, and everything in between, says Professor Michela Milano.

Europe needs to be autonomous in terms of artificial intelligence (AI) and create its own AI technology that respects citizens’ privacy, says Michela Milano, Professor at the University of Bologna and one of the leading AI researchers in Europe.

The European Commission wants European experts to build AI that people can trust, and according to EU’s ethical guidelines, trustful AI is lawful, ethical, and robust. As the Deputy President of European Association for Artificial Intelligence (EurAI), Professor Milano has a central role in this work and she supports the commission in shaping its AI strategies. She leads also the interdepartmental research institute for human-centred artificial intelligence, Alma AI, in Bologna, and in November, she will speak about the European AI strategies at AI Day 2020.

Europe needs to be able to compete against large economies, such as the US and China, in terms of AI. According to Milano, its advantages include the strong university system, the long history in conducting research of top quality, and an industrial ecosystem that includes start-ups, big corporations, and everything in between.

'We have all the components that are needed to create a significant strategy and to really create an autonomous Europe. But there are problems, too: investments and the fact that start-ups can start here but as soon as they become successful, they are bought by Google, Facebook, or another big American player,' Milano notes.

Another significant problem for Europe is brain drain. 'Talents are leaving Europe because they can find better salaries and better conditions outside of Europe. We really need to retain talent. If we have a strategy with an ecosystem that is favourable for them, and for creating start-ups and keeping the start-ups in Europe after they become successful, I believe Europe will have a very important and competitive advantage with respect to other big players.'

Taking privacy issues seriously is crucial. “For example in China, privacy is not considered important at all and personal data are shared and used without any consideration. Europe should keep on taking these values and aspects very carefully into account to make something really different.”

Building trustworthy AI is difficult but the right path

Building trustworthy AI is more complex than what it may sound. Milano points out that many AI systems “simply work”, while we do not complete understand why they work. To create trustworthy AI systems, researchers need to develop AI that is understandable and explainable to humans.

'We also need to have more collaboration between AI systems and experts, which is very difficult. That requires encapsulating the collaboration and interaction with an expert. The system needs to learn from humans how they solve problems, and that knowledge needs to be injected in the AI systems.'

Europe wants to shape a strategy that differentiates from other economies, like the US and China that have made huge investments on AI.

Michela Milano

When we are developing such systems, we also face the issue of causality; while calculating correlations is straightforward, showing causal relations – that outcome Y happens because of the factor X and these events do not simply co-occur – is much more demanding.

'There are many, many aspects that are considered important from the European point of view. Europe wants to shape a strategy that differentiates from other economies, like the US and China that have made huge investments on AI. Europe is lagging behind a bit, but I think that the strategy that is going to build trustworthy AI in response to ethical principles is a good path.'

Difficult questions call for interdisciplinary collaboration

Professor Milano has been an AI researcher for much longer than AI has been the hot topic it is these days. She remembers the time when people commonly thought that AI methods were useless and researchers even avoided using the term to avoid scaring funders away. 'But now of course everyone understands that it’s useful and can bring really important results and impacts in all aspects of our life.'

Milano researches decision-support systems and especially systems that support – rather than replace – human experts. What Milano thinks is particularly fascinating in AI is that researchers can embed human knowledge in AI models. 'You can really use these systems to put together knowledge from different experts, different domains, and glue it all together with data-driven models.'

One aspect that fascinates her is the interdisciplinary nature of AI and, for example, ethical aspects that call for multidisciplinary approach. 'When you are in a driverless car, you need to consider aspects that are not just algorithmic or technological. You really have to understand that there are difficult decisions to make.'

When e.g., lives are in question, modelling a system is particularly difficult and ethical and moral principles should come into place – and at the same time, there are many situations in which people from different parts of the world do not follow any global moral guidelines. While East Asians may think that it is most important to protect the elderly, Europeans put children’s safety first.

How to code these types of decisions will be one of the key challenges in AI research in the near future. On a positive note, a machine can be more rational than human decision makers can, as it can process information so quickly that it can genuinely base its decisions on reasoning, while humans use their instinct.

Technology helps to make the planet a better place

In the last few years, artificial intelligence systems have developed at an extremely fast pace. Professor Milano is eager to see what types of results will be achieved in the next ten years; she hopes to witness a European AI field that is truly strong and competitive.

Another thing she wishes to see is that researchers start using AI systems more to solve problems that help make the Earth a better place for everyone; technology can help us reduce pollution and emissions and hit the zero emission and carbon neutrality targets.

'It is time for us to use technology for the good and help the environment and improve our citizens’ quality of life. Technological solutions really help us to go towards a better planet for us, our children, and the next generations to come.'

AI Day 2020 takes place on 26 November 2020. Read more at

  • Published:
  • Updated:
URL copied!

Related news

ARTS open science roadshow, pic of the session
Research & Art Published:

European Open Science Cloud (EOSC) Finnish Forum webinar on 25th January

One of the speakers is Karel Luyben, the first president of the EOSC Association and a member of board at Aalto University.
Installation Talk 2020, Jarkko Niiranen
Research & Art Published:

Professor Jarkko Niiranen on the computational mechanics of microarchitectural solids and structures

'When developing models, we spend a lot of time on mathematical physics, numerical mathematics and computer programming, but the fundamental concepts of mechanics remain the same' says Niiranen in his installation talk.
A photo showing Dr. Dorothea Golze
Research & Art Published:

Computational physicist Dorothea Golze receives prestigious Emmy Noether Award

Dorothea Golze received funding from the German Research Foundation within the Emmy Noether Programme to establish her own junior research group at the Technical University of Dresden.
Dronen ottama kuva Otakaari 1:sestä, kuva: Mikko Raskinen
Research & Art Published:

How do you know where a drone is flying without a GPS signal?

In Jouko Kinnari's doctoral dissertation, the location of a drone can be determined using map data and sensors.