In an insightful conversation with Jost Dubacher, Dalith Steiger-Gablinger, co-founder of SwissCognitive, sheds light on the underestimated prowess of Switzerland’s AI landscape. She delves into the reasons behind Switzerland’s quiet yet impactful presence in world-class AI research.

 

Credit: This interview by Jost Dubacher with Dalith Steiger-Gablinger has been published in German and part of the article “Wir haben KI-Forschung auf Weltniveau” – “Interview with Dalith Steiger: AI, We Have World-Class AI Research In Switzerland”


 

SwissCognitive positions itself as a hub for the global AI community and offers consulting for investors as well as support for startups. Dalith Steiger-Gablinger, a mathematician, keynote speaker, and serial entrepreneur, co-founded the platform.

Since the launch of ChatGPT a year ago, everyone has been talking about AI. You have been in the field for longer. Why?

I have to dig a bit into my biography. After my studies, I went into software development at UBS, then I managed clients at the CIO level at Avaloq, and finally, I led the Swiss IT Leadership Forum, a consortium of executives from the Swiss ICT industry. So I learned many aspects of computer science when I heard a presentation on Digital Employee and Avatars at a Gartner event in 2013. It was a revelation to me. Since then, I know how powerful AI is. It changes our lives.

You advise international companies on AI strategy, among other things. How is Switzerland as an AI location perceived from the outside?

We are underestimated because we don’t sell ourselves well enough. AI research has been conducted in Switzerland at a world-class level for decades; I would like to mention Jürgen Schmidhuber from IDSIA in Lugano, one of the most cited computer scientists and AI experts ever, or Roland Siegwart from ETH Zurich, a globally recognized authority in robotics. We need to catch up on the business side in implementation.

Good keyword. According to evaluations by the Startup Radar team, AI applications play a business-critical role in almost one in four Swiss startups. Are you surprised?

Frankly, I thought there would be more. But let’s see how and where these companies are involved.

40 per cent of all AI startups optimize existing products and services with AI applications, and just as many use AI to develop new things.

I find this number encouraging. Because AI has enormous potential to accelerate existing processes, but the real power of self-learning algorithms is shown in pushing the boundaries of what’s possible – for example, in the field of drug, material, or substance development.


Thank you for reading this post, don't forget to subscribe to our AI NAVIGATOR!


 

Another 20 per cent of AI startups explicitly deal with the further development of models and algorithms. What do you think about that?

This number is also good news. It shows that founders have discovered the topic for themselves.

It’s noticeable that the focus of startups so far has clearly been on discriminatory AI / Deep Learning. Projects that deal with generative AI (GenAI) are rare. Does this match your perception?

Since I move exclusively in the AI ecosystem, I would have expected a higher number. But let’s talk again in a year! The chat function on GPT 3 and the launch of GPT 4 have given generative AI a tremendous boost. We see in the startup accelerator program of SwissCognitive that many GenAI projects are still in stealth mode. We’ll hear more from them in the coming months and years.

Countries like the UK are investing billions in AI research. Switzerland is holding back. What makes you confident that Switzerland can actually establish itself as a global AI hub?

Besides strong basic research, it is mainly the political aspect: AI, like any new technology, must be regulated. In Switzerland, we have the opportunity to create frameworks that are more practical and leaner than the regulatory frameworks in the EU or the USA. We succeeded in doing this with blockchain technology, and I hope we can repeat this success.

Interview: Jost Dubacher

Original article