GDI: From AI-generated images to intelligent contact management software: where does (social) AI begin, and where does it end?
Karin Frick: AI systems learn quickly and are becoming universal tools. What we do with them – whether we use robots for organic farming, for Mars missions or as hoovers – will soon be limited only by our imagination. Most AI applications run in the background, invisible to end users. Similarly to light, only a small fraction of the AI spectrum is directly visible to the user. What is new is that AI and systems such as ChatGPT now understand human language, you can talk to them directly and give them increasingly complex tasks that they can solve independently.
How is AI changing our relationships and us humans?
The ability to communicate in human language turns machine intelligence into a kind of (imaginary) person. AI can take on many roles: it can become our slave, coach, copilot, doctor, boss, friend or pet. In the future, anyone looking for help will probably ask AI first. At best, this will make us less dependent on experts or other people who know more and therefore have more power. At worst, it will make us feel increasingly powerless and manipulated by AI systems whose workings we do not understand.
Can AI improve our social relationships? And if so, how?
We can primarily delegate more tedious tasks to AI so that we have more time for direct communication with other people. For example, if a robot in a hospital takes over individual body care or room cleaning tasks, carers can use their time to talk to patients in person. On the other hand, a growing proportion of business relationships can be handled directly between one machine and another. In a sense, a shoe factory can automate the entire process, from the design to purchasing materials, contracts, production and marketing. In the future, everything that a machine can do will also be done by machines. People will concentrate on the areas where human skills are particularly in demand and will therefore perhaps need their social skills more rather than less.
Will we become more suspicious if AI is everywhere?
We already need AI every day when we search the internet, use a smartphone, pay digitally or drive an electric car. I suspect that, in the medium term, we will develop a similar relationship with AI as we do with electricity: nothing works without it, it is available as a matter of course, but we only pay attention to it when it fails.
What are examples of how AI influences our social interactions with people?
Artificial intelligence influences our social interactions with people in various ways. Firstly, it enables us to solve more questions on our own, making us less dependent on other people. Secondly, it can help to reduce information inequality. This makes it easier to negotiate on an equal footing, since all parties have similar information. One area where AI could have the greatest impact in the short term is communication between different languages. AI systems can translate simultaneously between different languages such as Chinese, German, Russian and Spanish. This opens up unimagined possibilities for direct communication worldwide by enabling people to get in touch with each other despite speaking different languages.
Which industries will be most affected?
AI affects all industries. In the context of the GDI Trend Day, B2C communication with AI chatbots is particularly interesting. On the one hand, more AI sales bots will try to make the shopping experience more personalised by tailoring answers to the language and mood of consumers, responding directly to questions and advising them on training and diets, for example. One important area of application is topics that require a relatively high level of advice, such as financial or nutritional advice, travel guides, etc. Wherever personalised advice was previously exclusive and available only to premium customers, it can now be offered to all customers.
Now let's look at the big picture: will AI make it easier to solve global problems in the future?
AI is a powerful universal tool, but not a panacea. AI is helping to advance research into many previously unresolved climate and health issues. People will continue to decide on the implementation and realisation of new findings in the future. AI will also advise politicians, but will not govern (for the time being).
Does AI make better decisions than politicians?
AI can be used to create a better basis for decision-making. It is less about AI making decisions instead of politicians and more about showing better ways of achieving a desired goal. For example, by showing alternative ways to reduce healthcare costs or regenerate biodiversity.
Is it cooperation or will AI surpass us?
AI can already do many things better than humans. AI will expand the range of what is conceivable and feasible in a similar way to the invention of the telescope or microscope. But it is also associated with risks that need to be regulated. The greater the risk, the more regulation is needed, just as with nuclear power and biotechnology, where there are very strict authorisation laws.
Will there be a marker for AI influence, similar to a watermark?
Transparency creates trust. Just as we have to be able to declare and prove the origin of food, in future we will also be able to prove the origin of images, text and videos. "Self-thought" will become a quality feature.
Learn more about AI at the 20th European Trend Day,which will be held at the Gottlieb Duttweiler Institute on 13 March 2024 under the heading "AI buddy: redefining social relationships". International AI pioneers will show how humans and artificial intelligence can work together to shape the future, they will present the latest findings from research and showcase the most exciting AI startups.