"Machine Learning 2021" in the subscription shop

In the lower places are factors such as a higher degree of innovation or the fact that new products and services have been created with the help of AI and machine learning. This is especially true for small and medium-sized companies. Around 40 percent of large companies, on the other hand, rate the success of an AI and ML project primarily on the basis of an increase in their innovative strength .

 The success of an ML project is primarily determined using classic parameters such as higher productivity and cost reductions.

This could be in arrears to the fact that, especially for medium-sized companies, strategic aspects play a lesser role when using both technologies - for example the option of entering new market and product segments. They prefer to reap "low-hanging fruit" first, such as cost reductions through greater streamlining of work processes through AI and machine learning.

However, it would be a mistake to disqualify this approach. Because withoutriskis not the commitment in the field of AI and machine learning . According to the study by IDG Research Services, ten percent of the companies surveyed have cut the existing budget for AI and ML. The reason for this is likely to be that such projects did not bring the desired success . This is why experts from providers of such solutions, such as Nvidia, Microsoft or Lufthansa Industry Solutions, advise you to start with less ambitious projects first and to gain experience with these technologies first.

To the study "Machine Learning 2021" in the subscription shop

New products and process automation

But the bottom line is that the positive signals predominate. So have 63 percent of the Companies developed a business model to use AI and machine learning to develop new products. And half see both approaches as a means of generating indirect added value, for example by optimizing and automating internal processes.

In the last-mentioned point in particular, process automation , there has been a change compared to previous years. The fear that algorithms will in a sense take over the command of business activities is gradually receding into the background. Obviously, the realization is gaining ground that, especially against the background of digitization, the speed of reaction and agility of the company must be increased. This is where artificial intelligence and appliance learning can make an important contribution.

One example are AI-based customer information systems that work on the self-service principle. Chatbots answer questions from customers, for example (30 percent of the use cases). And around 37 percent of the use cases aim to automate the processing of processes (such as damage reports). In this context, AI technologies such as natural language processing play a central role.

However, as in previous years, the most important field of application is IT(76 percent). The reason is that the increasingly complex IT and cloud environments can no longer be managed "manually" and protected from cyber attacks. Machine learning algorithms, for example, are able to identify the first signs of such attacks faster and more precisely than administrators and to automatically initiate countermeasures.