Why we need an ethical compass?

As National Technology Officer and member of the Executive Board of Microsoft Switzerland, Marc Holitscher supports selected customers in the implementation of innovative business models. In particular, he works closely with companies in the financial industry and supports them in the holistic assessment of relevant opportunities and risks in the adoption of cloud-based scenarios. In addition, Marc Holitscher is responsible for positioning strategic topics such as cybersecurity or artificial intelligence with decision-makers in the commercial and public sectors.

Mr. Holitscher, what do you understand by artificial intelligence (AI)?

Artificial intelligence (AI) is a so-called enabling technology. This means that it forms the foundation for further innovative steps - and this across all areas of life and the economy. The term AI itself describes various technologies that work together to complement and strengthen human abilities in seeing, hearing, analysing, deciding and acting. It's not necessarily new, but three factors are accelerating the development of AI: big data, the cloud, and powerful algorithms. AI is meant to support humans, not replace them. Our goal is therefore for humans and machines to work together in the best possible way.

Which industries are particularly challenged by AI today?

There is no sector that is not being transformed by AI. We explored this question in a joint study with EY: Especially in the life sciences sector, 96 percent of all executives expect AI to have a high or very high impact. But expectations are also high in the service sector, and similarly in finance. Analogous to the companies, however, national economies must also prepare themselves. This also applies to Switzerland. Across all sectors, executives expect positive impulses for customer service, product innovation as well as improved efficiency of their companies and for more attractive working conditions.

AI contains opportunities for the future - where do you see the threats (from a company and society perspective)?

We are aware of the tension between the opportunities and challenges of AI. Our starting point is: People must be at the center. Technology must be designed to complement and extend human ingenuity and capabilities. This is precisely why it's so important to be proactive about AI and to be able to classify it in its shades. This is as true for businesses as it is for individuals. Is everything that is technically feasible also commercially and socially desirable? Consider, for example, the discriminatory effect of AI when it is built on a skewed data base. We are seeing more and more companies asking these questions and looking for principle-based ways to develop their own risk management.

There are great reservations and fears in society today about AI - how can you gain the consumer's trust?

The absolute prerequisite for the human-oriented use of AI is that it is carried out according to clearly defined values and rules. Microsoft has defined six ethical principles for this purpose. They are: Fairness, Reliability, Privacy and Security, Inclusivity, Transparency and Accountability. Transparency is of central importance here: people must understand how a decision is made and what criteria were used to reach it. We are currently investing a lot in this aspect of explainability.

What do you recommend to companies and society (economy, politics) with regard to dealing with the dangers?

Personally, I regret that public discourse on this forward-looking topic is practically non-existent in Switzerland. Now is the time to make conscious decisions regarding the framework conditions for the development and use of AI. It is true that there is an Artificial Intelligence Working Group, which reports to the State Secretariat for Education, Research and Innovation. However, this is interdepartmental and there is insufficient exchange with the private sector or civil society.

Dialogue and knowledge transfer is particularly important for our country because Switzerland is inherently well positioned with its leading universities and universities of applied sciences and as a location for emerging AI companies. We should promote more intensive exchange, because a lack of awareness of the application and impact of the technology is one of the biggest challenges.

As a major global player, how do you deal with the issue of AI and ethics at Microsoft? How do you take your responsibility in this context?

We have developed six principles that we believe should guide the development and deployment of artificial intelligence. I have already mentioned these. Crucially, these are not lip service. There are concrete processes and controls in place that ensure these principles are implemented in our organization. For example, we always evaluate AI projects to determine whether they are ethical or not. In the process, we may also decide not to conclude the deal.

In addition, we want to further advance the social debate on these issues. This includes, for example, the discussion of the areas of application in which we want to explicitly regulate the use of artificial intelligence. Facial recognition is repeatedly cited as an example here.