Thomas Jenewein, Business Development Manager at SAP, explains how the software company trains its employees in dealing with AI and why experience-based learning is the key to success.
Working and learning with AI represents one of today's major trends. Beyond the opportunities, there are also challenges, such as establishing the technology among employees and complying with regulatory requirements like the EU AI Act, which obligates companies to ensure AI competency among their staff.
As part of the "AI-Humanitas: Enabling AI-Skills" initiative, Dr. Wolfgang König interviewed Thomas Jenewein, who is responsible for business development for training, adoption services at SAP and hosts the EducationNewsCast podcast.
The Decoder: Mr. Jenewein, what are the three most important key learnings from your perspective on the topic of working and learning with AI?
Thomas Jenewein: Off the top of my head, I'd say everyone needs to experience it themselves. I've learned the most through hands-on application. The technology changes constantly – recently, Perplexity introduced the ability to work with DeepSeek and AI reasoning models. The biggest change in my work isn't necessarily completing tasks faster, but taking on additional tasks, like summarizing podcasts. Working less and having more free time doesn't happen. Instead, the quality and output of the work increases, along with my creativity – I wouldn't come up with many ideas otherwise.
The Decoder: Is artificial intelligence already being used at SAP?
Thomas Jenewein: Our main focus is on developing AI products – embedded AI, our own chatbot, and a technical platform that integrates and extends AI as part of our ERP system. That's certainly our greatest potential. Of course, I estimate about 70% of our developers use GitHub Copilot to improve their work and become more productive. We also have an area called AI-Launchpad, where about 30 language models are available for both developers and employees to use. Everyone starts experimenting within their job function to see if it makes them more effective and productive.
As a knowledge worker, I use AI myself – Microsoft Copilot, Claude, and various language models for podcast production. These are part of my business development responsibilities, especially when building an external community. I use them where it makes sense, such as for translations, transcriptions, preparation, follow-up, summarization, and writing texts like blogs. Information gathering and research is also a major area where I naturally use AI tools like Perplexity.
The Decoder: How exactly does SAP empower its employees?
Thomas Jenewein: We have various educational initiatives and programs designed to promote understanding and application of AI. These range from online courses and workshops to internal training specifically aimed at familiarizing our employees with the latest AI technologies and practices. The goal is to build a broad base of AI competencies that enable SAP to remain at the forefront of innovation as a leading technology company. The approaches vary depending on job function.
You might be familiar with our Promptathon Sessions format, which lasts two hours and has been very well received. Thousands of employees have already learned how to use AI tools hands-on through these sessions. For developers, the approach is different – they receive specialized training, particularly for GitHub Copilot. Studies show that developers gain the most in productivity, followed by areas like technical writing and marketing. There are some formal training programs but also many informal learning opportunities. I'm a member of two innovator communities, such as the Experience Garage, where regular sessions on topics like ethics or new technologies like Agentic AI take place. These serve as networking events and provide brief insights. There's also a dedicated AI community.
I consider myself a "champion" in our AI community, especially in the use and promotion of AI within our organization, such as with Microsoft Copilot. It's crucial that every employee has the opportunity to experiment with AI and learn how to best integrate it into their daily work. This ranges from developing new products to improving everyday work processes.
The Decoder: That sounds like a comprehensive strategy to keep pace with current technology trends.
Thomas Jenewein: It's important for us that all employees have access to the necessary resources and training to ensure they not only understand AI technology but can also use it effectively in their respective roles.
The Decoder: How does SAP deal with the EU AI Act?
Thomas Jenewein: At SAP, we've developed a comprehensive Responsible AI Framework that ensures our AI applications are designed and implemented responsibly. Beyond compliance and security, AI ethics is very relevant. For example, we've established a highly structured process supported by our ethics assessment and a specialized committee. This committee evaluates cases, particularly "Red Light" use cases that might raise critical ethical concerns, such as processing personal data without human intervention. The goal is to ensure all AI applications function transparently, traceably, and in accordance with established ethical standards.
Additionally, SAP has extensive experience in applying AI and machine learning. Our ethical regulations were recently adapted to UNESCO guidelines, requiring only minimal adjustments since we already had high standards embedded in our processes. To promote awareness and understanding of these important topics, we offer a free e-learning program accessible through our website. This program is available to everyone without registration and includes videos and quizzes focused on our AI ethics framework.
The Decoder: What about acceptance of AI technologies? Are there any resistances or challenges you need to overcome?
Thomas Jenewein: Of course, there are always challenges when introducing new technologies, especially disruptive ones like AI – whether it's fear of job or relevance loss, or uncertainty about how to use AI. In modern terms, this topic can be described as AI adoption. It shows that change management, empowerment, technical support, and a human-centered approach are fundamental.
The Decoder: In conclusion, what do you consider most important about working and learning with AI?
Thomas Jenewein: The concept we've long advocated is experience-based learning. AI is a tool, and like any tool, it's often generic. The real challenge lies in transforming your own work: exploring what can change, what you should do more or less of. I'm convinced this can only be achieved through experience-based and experimental learning. Of course, there are global use cases like software development that represent a significant lever for us. But as knowledge workers, where creativity is required, everyone must discover what works for themselves. Exchanging ideas with others is essential. This isn't new; we've always emphasized this in the corporate learning community, but now it's more important than ever.
