Hello, and welcome to our Analyst newsletter today. I will talk about AI in a nutshell, what AI is and what not. So AI probably is the password of the year, these days. So what is it really? What is it not? I trust one quickly walk through these passwords. So in a broad definition, artificial intelligence or AI is the science of making computer soft tasks that usually require human intelligence. And so the most granular level where we can split us between strong AI or general AI, which is really very computer, sort of has to same minded, exactly the same sense human beings have mind. We are far away from that. So let's skip that. Let's create out and move to the things which are more reality, which is weak AI, not a friendly term or applied AI. So software, which is focused on solving specific problems.
And that is where we have seen a lot of progress over the last years and where we expect to see further broker over the next couple of years. So here we are looking at obviously the ongoing AI research. So what is versus is happening. And there are a lot of fields which are really concrete cognitive solutions. So from, from our definition, cognitive solutions are practical applications of AI research, things that you really can use in practice. And there, we have a couple of different levels we should look at and a couple of areas where we can use it. So we see cognitive solutions in many, many fields. We see it in information security, identifying anomalies and outliers self-driving vehicles, very popular application, psychological profiling might be a little scary postal mail address detection, something where we see this for quite a while. And many, many more.
We see more and more things, augmented capabilities in production, environments, cetera, et cetera, these solutions. So the practical applications are based on cognitive technologies, technologies such as computer vision or language processing. So speech to text, text, to speech, knowledge, representation, and many others. And so these technologies, again, build on various types of frequently machine learning methods, not only, but to a large extent. And when we look at this machine learning methods, we see things like pattern recognition. So really recognizing a pattern correctly, learning about patterns and getting better and better outlier detections. So what are the anomalies, genetic algorithms, deep learning and others, and there's would be a lot to talk about what are the different technologies in specific, but basically it's really just solutions to technologies,
To machine learning methods, to underline algorithms and methods such as neural networks, such as cluster analyzes, such as regression analyzes and more. But we also have a couple of areas which are not really AI, which are just, which is still sometimes quite complex and very helpful, but trust sort of statistics and other stuff. So net machine learnings, everything which is only based on redefined rules and several other stuff. But if it's just rules, it's not machine learning until the machine learns and is trained. And part of machine learning usually is really also training element. It's really that machine learning and it's not AI, there are other algorithms and methods. So pattern matching more on a statistical basis instead of learning about patterns and recognizing them better and better over time. So this pattern matching thing is really more a statistical thing which can move in an advanced incarnation into pattern recognition, really into the field of AI and many more factually, most of these cognitive solutions are a combination of different cognitive technologies plus traditional technologies.
And that's always the point combine it the right way. So what I strongly believe is that AI is something which is done right? And with a good perspective on ethics and governance, something which is very valuable to our businesses and to the entire society, we need to be careful, but there's a huge potential in helping people doing their jobs better. That is what AI is really doing, helping people to do their better. Augment the capabilities of people. Look at edit from that perspective and spend time on analyzing what is happening in AI. We will deliver a lot of research over the next year around AI, have a look at our research. Thank you.
