"The need of man does not disappear completely"

- It's a pretty misunderstood technology. AI is not about humanoid robots running on the street, says Peter Sarlin, professor at Hanken's School of Economics in Helsinki and founder of AI company Silo AI.

It's far from the level of "super-intelligence" when AI can think for itself and face new problems in a creative way, he says. Today it is a narrow use where people have to feed the AI ​​with the right data in order to learn how to distinguish a cat from a dog. Or identify a person in the picture. But artificial intelligence is already raising concerns about its application.

- You can do damage with algorithms if used unethically, Sarlin notes.

The debate on AI has been very much about which jobs will disappear when more and more companies automate production with the help of technology. Peter Sarlin gives a typical example of that.

- In many factories, there are still quite a few people sitting behind a screen or standing by the running belt and monitoring the quality of what is produced, this is something that machines can usually do just as well or better.

But he doesn't think the AI ​​robots will take over completely in such workplaces.

- It is not so that the need for man disappears completely. But man's duties change. It's nothing new when it comes to technology issues.

The fears are also about personal privacy. Artificial intelligence enables, among other things, facial recognition, which is used to unlock the phone or pay with the face, to identify suspected criminals on surveillance cameras and chase dissidents in authoritarian regimes.

Recently, the New York Times revealed how a small American company, Clearview AI, vacuumed the net on three billion images taken from social media and used to identify, for example, who is stuck in the image of a surveillance camera.

Where does the limit go when you infringe on personal integrity? Today, the European Commission is expected to present a so-called "white paper" with suggestions on how to best use and set limits on the use of artificial intelligence, so that authorities and companies that are today very eager to apply it will have a framework to adhere to.

However, Peter Sarlin, whose company works for companies that want to use AI in their operations, thinks that one should beware of painting any terror scenario.

- As long as we guarantee that we know what we do, what data we input, we know what algorithms we build, and we know how to use it in the end, then I think everything is okay.