AI - anything but rocket science

When people hear the term “artificial intelligence”, many people think of science fiction or Hollywood blockbusters in which robots take over the world. But artificial intelligence (AI) has long since arrived in our reality - and not at all threatening, but extremely helpful. In our “Artificial Intelligence” series of articles we want to show which application scenarios are already being implemented with AI and how companies can benefit from intelligent technology. In this post you will learn why today's AI services are by no means rocket science and what possibilities the interaction of AI and ERP system offers your company.

A wide field

Experts still do not agree on an exact definition due to the large area of application. This is mainly due to the fact that the term “intelligence” is also not precisely defined. In research, however, a fundamental distinction is made between weak and strong AI, which are differently “intelligent”: With weak AI, algorithms answer questions whose answer they have previously learned. We encounter this form of AI everywhere in everyday life - for example in speech recognition or in the form of product recommendations when shopping online. So far, strong AI, i.e. algorithms that develop understanding and awareness, has only existed in science fiction.

The approach: from science fiction to the present

Nevertheless, AI can already do a lot today - even if your own thinking is not (yet) part of it: Microsoft understands AI as “technologies that complement and strengthen human abilities in seeing, hearing, analyzing, making decisions and acting.” According to the Fraunhofer Institute, AI includes everyone “IT solutions and methods that complete tasks independently, whereby the rules on which the processing is based are not explicitly specified by humans.” The last part in particular is particularly exciting: Thanks to certain algorithms, machines gradually learn to take on certain tasks and to improve.

A journey through time: the most important AI milestones

1950: Development of the first computer intelligence test ("Turing test").
1956: First use of the term "Artificial Intelligence" (Eng .: "Artificial Intelligence").
1966: The first chatbot goes live.
1997: A computer defeats the world chess champion Garry Kasparov.
2011: The first personal language assistant is used.

Please feed with data

The prerequisite for this ability to learn is that the machines are supplied with sufficient data. Based on repetitions, they can then recognize data structures and independently perform a previously defined task - keyword: machine learning. However, when it comes to feeding the data, machines are still dependent on us humans. Data scientists ensure that the information provided is of high quality and reliable. Another important and often neglected requirement is that there is sufficient storage capacity. Otherwise important data will be lost.

Dream team: ERP system and AI

But where does the data come from? For example from the ERP system. As a central data hub, all information comes together here in many medium-sized companies. By using AI services, companies can then automate processes, improve security and gain insights more quickly. Before doing this, however, companies should be clear about what they want to use AI for. Solutions that have already been implemented by other companies and serve as a blueprint for your own projects can be helpful here. These can then be specifically adapted and trained with your own data. With Microsoft Azure, for example, prepared training data can be uploaded and evaluated in the Azure Machine Learning Studio. The algorithm trains itself with part of the data and can then analyze further data independently. Because many AI models can now be trained online and then used offline, companies are also on the safe side when it comes to data protection.

What are you waiting for?

The powerful Microsoft AI platform gives all companies the opportunity to benefit from AI - regardless of whether they are large corporations or family-run companies. And the best is: ERP solutions based on Microsoft Dynamics 365 work seamlessly with the AI platform. So what are you waiting for? Anyone who already connects AI services to their ERP system can secure a clear competitive advantage. Because machine learning helps, among other things, to better evaluate data and automate processes. In addition, AI provides answers to complex questions and ensures that documentation obligations are reliably fulfilled - even if several people are involved in the project. For example, AI is used in the healthcare sector to facilitate and accelerate documentation using speech recognition. Would you like to make your ERP system fit for the future? Then contact our digitization experts, who will be happy to provide you with information on how you can implement AI in your ERP software: https://kumavision.com/kuenstliche-intelligenz