AI in Focus—Between Hype and Reality

The hype surrounding AI has waned as quickly as it emerged. Is the hype over, or has AI simply become part of our everyday lives?

Does it really matter?

Nowadays, an AI headline or a "powered by AI" stamp in an app is no longer enough to lure anyone in. In the initial euphoria, it quickly became clear that artificial intelligence was here to stay. At the same time, it became clear that many of the supposedly promising applications were not as efficient as initially thought. In Europe and Germany in particular, data protection experts are focusing on companies that are experimenting with AI. For SAP users, this means that data protection and compliance are top priorities when implementing AI solutions.

Artificial intelligence is already being used extensively in other areas. It is standard practice on LinkedIn to include emoticons and individual images in posts. However, these are not the areas in which companies can really leverage AI to scale up and enhance the efficiency of their core processes. When AI is integrated into established processes, the challenges become evident. The lack of clear communication regarding process flows hinders the generation of creative ideas. It is not uncommon for adaptations to be overlooked in the course of a growing ERP system. Ultimately, the results frequently fail to meet expectations, and users often encounter tasks that cannot be solved within the system. Even when substantial sums are allocated to AI, the results are often disappointing. Models must be recalculated on numerous occasions, which is a costly process. Software must be partially rewritten, and several machine learning and deep learning models must be incorporated into the general LLMs to enable the swift, effective, and dependable resolution of specific tasks.

What will happen next? Does this indicate that early investments in AI were misguided? I do not believe this is the case. Identifying and addressing initial challenges ensures that promising concepts are rapidly transformed into valuable products that benefit the entire organization. Furthermore, early investments in technology provide a competitive advantage and persuade even those who are skeptical of technology to reconsider. For SAP users, this means they can take advantage of the latest AI developments and further optimize their business processes. The use of AI is enabling us to work more efficiently, with the ability to enter documents 100 times faster and utilize resources in production to greater effect.

Evaluation of AI decisions

One of the key challenges in the coming years will be to integrate and connect the individual areas that have been enhanced by AI. The governance of data and the verifiability of AI-generated decisions and data will become crucial tasks, particularly in critical areas. The quality of the data stored in the company, as well as its structure and linking, will be of significant consequence in this regard. This will allow us to fully leverage the potential of AI in all its forms and to assess and evaluate the results. This is a pressing requirement, given the significant costs associated with AI. Integrating the various AI-supported products remains a complex undertaking. It is often the case that project plans are closely defined, and yet defined processes still fail in individual cases, requiring constant adaptation and improvement. As the complexity of technology continues to grow, the importance of comprehensive external consulting will increase, particularly in areas outside the company's key strengths.

It is evident that we are now capable of surmounting obstacles that would have previously required human intervention due to limitations in computing power. It is evident that we will continue to make numerous decisions independently in the near future, but with the assistance of AI. To answer the following question most effectively, it is essential to consider the following: is AI simply an accelerated evolution of existing challenges, or is it the transformative force it has been touted to be?

To the partner entries: