{"id":9090,"date":"2025-04-22T09:20:23","date_gmt":"2025-04-22T09:20:23","guid":{"rendered":"https:\/\/icv-controlling.com\/?post_type=blog&p=9090"},"modified":"2025-04-24T17:33:03","modified_gmt":"2025-04-24T17:33:03","slug":"how-progress-comes-about-when-professionals-exchange-ideas-copy-copy","status":"publish","type":"blog","link":"https:\/\/icv-controlling.com\/en\/blog\/how-progress-comes-about-when-professionals-exchange-ideas-copy-copy\/","title":{"rendered":"WG Westfalen – Workshop launch with a full house"},"content":{"rendered":"
\n
\n
\n <\/div>\n <\/div>\n
\n
\"AK<\/div> \n <\/div>\n <\/section>
\n
\n
\n
\n

Workshop launch with a full house<\/h2> <\/div>\n <\/div>\n
\n

In line with the ICV motto “From practice – for practice”, the ICV Work Group Westfalen organized an online workshop on the basics of AI and ChatGPT in controlling – designed and conducted by the Work Group member Hans-Peter Schreiber. All 20 registered members from the DACH region actively participated. <\/p>\n

The initial focus was on the technological classification of AI and its increasing relevance – not only in business and science, but also in everyday life. The interactive parts, in which participants were able to experience for themselves how AI models are trained and “learn”, were particularly well received. <\/p>\n

AI basics with ChatGPT – A workshop for controllers <\/strong><\/p>\n

Artificial intelligence (AI) is no longer a dream of the future – it is already changing the role of controllers today. In the ICV workshop “AI basics with ChatGPT for controllers”, participants were given a practical introduction to a complex topic – understandable, interactive and with direct relevance to everyday working life. <\/p>\n

What is AI – and why does it affect us all?<\/strong><\/p>\n

After a brief definition of the term, it became clear that AI is a branch of computer science with growing relevance for controlling. Why? Because controllers work on a data-driven basis every day – this is precisely where AI unfolds its potential: recognizing patterns, generating texts, preparing reports, creating visualizations. <\/p>\n

How AI works – a look inside<\/strong><\/p>\n

It was clearly explained how an AI works with so-called tokens. In small exercises, the participants learned how language models “think” and react. The insight: what looks like “magic” follows mathematical and statistical rules – which can be understood. Current research (including Google) shows that AI will provide creative services by 2030 – with all the opportunities and risks. <\/p>\n

Legal framework: EU AI Act & ISO 42001<\/strong><\/p>\n

AI must be used responsibly. Since June 2024, the EU AI Act has been the world’s first comprehensive AI law – binding in all EU member states. Objective: to protect fundamental rights, security and the rule of law. Violations can be sanctioned with up to \u20ac35 million or 7% of global turnover. This is supplemented by ISO\/IEC 42001:2023 – the first international standard for AI management systems (AIMS). Particularly important for controllers: the responsibility of management in accordance with the GmbHG, AktG and the role of risk controlling. <\/p>\n

Prompting for beginners – the key to using AI<\/strong><\/p>\n

Prompting – i.e. the correct way to “address” an AI – was explained using practical examples. How do I formulate a question so that the answer is useful for my reporting? How do I optimize the results? Particularly impressive: the targeted control of the answer quality through the so-called “temperature” – from factual and precise to creative and narrative. <\/p>\n

In exercises, the participants created IBCS-compliant bar charts, among other things – directly from the AI. Another surprise was that ChatGPT can generate and improve its own prompts – a real added value for day-to-day work in controlling.<\/p>\n

What did the participants take away with them?<\/strong><\/p>\n

The most important insight: AI is not an oracle – it is a tool.<\/strong> And the better you understand it, the more useful it becomes. Our controllers went home with the following learnings: <\/p>\n