Next Gen AI, what can we expect?

What’s up next for Artificial Intelligence. We made good progress on narrow AI, very good at specific tasks like object recognition and speech to text. On the other hand, we are still far off from General AI. Various challenges still exist in transparency of the outcomes of AI, trust by the ones using the algorithms and the shortage of skills. AI will play an important role to overcome these shortages.

Join us in our chat with Wouter.

Yves Mulkers
Hi and welcome, today we’re together with Wouter Denayer, Chief Technology Officer at IBM. Wouter, you’re kind of authority in Belgium and I think outside the borders of Belgium as well on artificial intelligence.

Last 30 years. where I look back on how I got education, and how things are evolving. We see now with the online courses like Udemy, there is more on microlearning, our attention span is going down. There are things changing in there. It’s hard when you see the mismatch. Even with my kids, I see a very strong mismatch. with how traditional schooling goes and how they want to consume information and knowledge. Do you see an evolution, how it could improve our typical traditional learning?

Wouter Denayer
It’s true, while it may very well have started now, with the quarantaine we are in. Everybody’s learning online also the kids. My daughter is at university, all the lessons are recorded, and she can follow them whenever she wants on-demand, there is no classroom anymore. So why not? I think that a lot of sectors where this was thought to be impossible or unacceptable. We have kind of broken the door open now and we will see changes where you maybe learn the lesson before and then you have kind of contact moment with your professor where you ask questions and do group work, which can also be online. Where you really work together but the first processing of the basic information of the lesson, why would you have to be in a classroom? It’s actually not necessary and not even useful. This will change and we will see a clear acceleration of this because many people see now that it’s possible. Because they’re forced, We are creatures of habit and there was resistance. And I think now, while some of the resistance is gone, and people see this is actually useful, this is working. I’m really optimistic about that.

Yves Mulkers
You say it, you’re very optimistic. If you look five years further in the future, how do you see the development related to artificial intelligence and technology, I saw a lot of technology on waste recycling as well where IBM is working on, I feel confident that everybody’s still positive. I did recently a survey to my audience, and although there is a bad situation, people don’t have food, people don’t have financial security. They’re still pretty positive. For me, the future looks bright. What if you could look into the future for the next five years what’s going to come?

Wouter Denayer
Predictions are difficult. I’m optimistic about the role that AI can play in maybe not solving, but helping us address some of the big problems in society. If you take this example now that we have people with COVID-19, and the role that AI and high-performance computing is already playing and trying to find a cure is quite amazing, the AI systems, for example, are being used to try to accelerate the discovery of certain molecules that could be turned potentially to drugs that target certain aspects of this virus. We have AI systems reading all literature that is available which nobody can do. There’s an explosion of content on COVID-19, nobody can read all that. If you have a system an AI that can try to figure out what maybe are commonalities between all these papers, what is, for example, hidden knowledge. Hidden knowledge that can only be understood if you really like 10 of these papers, and then you have to make the link. What if the AI can make the link for you? It will not give you the solution, it will tell you this could be interesting, have a closer look? It is also used to create virtually new molecules, where we think that they could have an aspect again, helping scientists not waste their time. We know drug development is extremely expensive. So how do you help scientists focus? Not by reading patents. So I read in the literature, by trying to figure out where the interesting areas are for human exploration. We are really getting into this human-machine partnership mode, where we understand that together, we’re stronger

There is no fear because it is really just helping us be more efficient and not wasting our time. It’s like what Kasparov saw when he first lost to Deep Blue. He was very disappointed, obviously. Then he saw, we’re getting into this new era, where humans and machines in that case, in chess, are stronger than machines by themselves. Now we’re seeing this in all other domains. I’m very hopeful for chemistry, material discovery. It’ s really quite interesting in finding new batteries. IBM just announced that we are closer now to a less polluting type of battery, which was also discovered using AI. Recycling programs, it’s everywhere, farming, precision agriculture being more efficient with the soil with the water using less pesticides. It’s really really promising, the things we’re able to do and frankly, we can use all the help we can get.

Yves Mulkers
I have one final take for your Wouter, to close up with some music most of the time. What is your favorite type of music? Or do you have a favorite band? because music is what connects us all, besides data, and AI and neural networks?

Wouter Denayer
That’s an interesting question. I didn’t see that one coming Yves, thank you for the question. I’m a big fan of music in my own language, in Dutch of course English as well. You know, I majored in English literature, which is quiet interesting for my current position you can get into some other points. Music in your language, hits you harder or or stronger than in any other language. I follow people like Bart Peeters or Raymond van het Groenewoud, which I really appreciate as well, I’m omnivore, going from classical music to techno in my car, but I do enjoy music.

Yves Mulkers
That’s a great one, Wouter thanks very much for the insights in artificial intelligence and that it will support us in a very positive and hopeful future so thanks again and see you soon

Wouter Denayer
Thanks Yves, My pleasure