~5 min read

The reality of AI: between hype and challenges

Artificial intelligence (AI) is a trending topic that is dominating the tech world. Companies such as Microsoft, Salesforce and Alphabet are investing heavily in advanced AI solutions. However, despite the hype and promising prospects, there are still some realities and challenges to overcome.

The current situation: hardly any AI use due to a lack of concepts

Amidst the enthusiasm for AI, a sobering reality is revealed in some areas. In Germany, 87% of companies consider AI solutions to be important in the next five years. However, most of them face problems when it comes to implementing AI solutions. This is because some companies have not yet developed a mature concept for the successful integration of AI solutions, which often require not only technical adjustments but also a redefinition of business processes.

The biggest cost items are usually the recruitment of experts, the search for suitable software and updating the technical infrastructure. The need to continuously maintain and evaluate AI models and keep pace with rapid developments in the field of deep learning is also underestimated. On the one hand, this applies to the operational areas in which AI solutions are to be used, but also has an impact on areas such as IT security.

Last but not least, it is important to consider the future challenges arising from the lack of a legal framework, which many countries and regions are still working on. The EU's AI regulation, which is expected this year, will certainly bring new challenges for companies.

Many companies therefore either do not (yet) have sufficient expertise or the necessary budgets to scale AI projects independently, or are still reluctant to integrate existing solutions into their day-to-day operations.

The lack of feasibility of existing AI models for companies raises important questions that overshadow the euphoria surrounding AI. It is becoming clear that the supposed ease of AI integration does not always correspond to reality and that companies face considerable challenges in turning the promises of AI into reality.

It doesn't always have to be AI – arguments for semi-automated solutions

Automated AI solutions such as Salesforce Einstein and, in some cases, certain "Copilot" systems offer companies a comprehensive AI platform integrated into their systems that automatically analyses data, creates forecasts and, in some cases, makes independent decisions to optimize business processes. These solutions are particularly suitable for complex or data-intensive tasks.

In contrast, semi-automated solutions rely on a symbiosis of human expertise and machine efficiency. They support employees in the decision-making process by providing suggestions or analyses, but leave the final decision to humans. This type of solution is particularly beneficial when human judgment and creativity are required or when AI is used in areas where ethical or legal considerations play a role.

Semi-automated solutions can be an interim solution, especially for companies whose processes are not yet ready for full AI solutions or do not currently allow them. Not only do they offer faster implementation, they are also less complex than full AI systems. Particularly in industries that have not yet been penetrated by the AI revolution, semi-automated solutions enable a pragmatic approach that brings companies benefits in a timely manner without rushing into a hasty AI deployment.

Companies can gain various advantages by using partially automated solutions:

  • Faster implementation: semi-automated solutions can often be implemented more quickly. Companies that do not want to wait for market maturity or regulations for certain AI solutions can quickly achieve efficiency gains by using semi-automated systems.

  • Reduced complexity: AI solutions can be too complex for some companies, especially if they do not have sufficient resources or expertise. Semi-automated solutions offer a less complex alternative that is easier to understand and master.

  • Flexibility and adaptability: Semi-automated solutions are often more flexible and easier to customize to a company's specific requirements. This enables a more precise focus on individual needs and processes.

  • Consideration of ethical considerations: The use of AI increasingly raises ethical issues. Semi-automated solutions can be easier to monitor and control in this context, making companies better able to comply with ethical standards.

The mix as a solution

The discussion about AI should not be conducted in a black and white pattern. The reality for now often lies in the mix of advanced AI solutions and proven, semi-automated approaches. Companies should carefully consider their individual needs and capabilities and pursue a strategy that considers both innovation and practicality.

The future of business will undoubtedly be shaped by AI. But while we wait for the final breakthrough, semi-automated solutions can offer significant added value. A balanced approach that takes advantage of current opportunities and waits for promising developments could be the key to sustainable success.

microsoft teams
microsoft teams cpre platinum partner
Dr. Glinz COVIS GmbH
Heerdter Sandberg 32
40549 Düsseldorf Germany
Phone +49 (0) 211 - 55726-0
Fax +49 (0) 211 - 55726-26
© 2023-24 All rights reserved