Artificial intelligence - often called AI - refers to a wide range of technologies that enable machines to recognize, act and learn. And in a degree as it was unthinkable just a few years ago. AI will change business models in a way that we have not seen since the introduction of enterprise IT systems in the 1980s.
We do not want to bore you with a full assessment of the topic at this point. On the one hand the connections are highly complex, on the other hand the current hype topics AI, Machine Learning and Big Data are difficult to distinguish from each other. Moreover, the application with new frameworks and cloud infrastructure is getting easier and cheaper. This makes the technology basically tangible for almost every software engineer and the barriers to entry in the topic are constantly falling.
Partial Autonomous Driving
Lane departure warning, glare-free high beam, pedestrian recognition - almost no car manufacturer dares to introduce a new model without at least some of the above assistance systems. Pedestrian recognition, for example, uses computer vision, i.e. machine vision to detect people in front of the vehicle. All have in common that you work with methods of artificial intelligence and thus make driving a little safer today. Little attention is focused on Fleet Learning, which is collecting data from every single vehicle on the road. Only with these huge amounts of data - which incidentally all lie in the data centers of the large Silicon Valley Group - is the next step towards fully autonomous driving possible.
The "old hands" knew it: Each machine in the factory has its peculiarities and if you fill in the right place in the right place oil, tightened a screw or just regularly off and on again you save yourself downtime or can keep the next maintenance short.
Today, experienced employees are either hard to get or retired. IT predictive maintenance systems are considered to be the most tangible aspect of Industry 4.0, using vast quantities of sensor data to improve productivity and prevent lost production
NLP - Processing of texts
Text is the most important form of human communication. With a text, it is possible to bridge space (e-mail) and time (book) and to keep information easily accessible. Thanks to the Internet, texts are available in huge quantities today and are just waiting to be exploited using Natural Language Processing. Last but not least, visualizations generated from texts help to understand complex relationships.
"Predictions are difficult, especially if they affect the future."
No matter if you attribute this quote to Karl Valentin, Winston Churchill or Niels Bohr: In any case it is worth knowing the possibilities and looking for advantages for your own company processes.
Fully Autonomous Driving
Let's be honest: standing alone would help us. Traffic jam assistants who drive us autonomously and safely through the morning traffic collapse will be able to use transfer times more productively. The prerequisite for this is not the next development step towards still cognitive abilities but also 5G networks. Only with de facto real-time communication can the data gained by Fleet learning be used in every driving situation.
One may stand by what one wants: Anthropologists see such machines as having a major impact on society. Future generations of sexrobots will have a much stronger similarity to human partners with artificial intelligence. And their communication skills are expected to make them more and more like interpersonal interaction.
Anyone who has ever tried Alexa to learn details about a product or to coordinate an appointment with Siri knows it: There's something else to come. Cortana & Co. are only beginning to exploit the potential of artificial intelligence. Here, considerable improvements can be expected in the coming years. A special attraction would be an independent solution from the big Silicon Valley corporations, which is available for everyone as an open source solution.