SMEs need more support when it comes to AI
Swiss SMEs need more support from public and private entities when using AI, as well as a clear regulatory framework. A study conducted by HWZ in collaboration with Swisscom highlights the progress companies are making in AI and the challenges they must overcome.
Based on the question of how far the introduction of AI has effectively progressed in Swiss SMEs, the study results show that 38% of the companies surveyed have already introduced generative AI or are actively considering its use. According to the study, typical areas of application for GenAI applications are HWZ Report content creation, marketing and process automation. Generative AI is followed by the use of data analysis at 35%, while machine learning lags behind at just 12%, according to the results. The reasons for this: a lack of understanding of the potential of machine learning and its perceived complexity.
Challenges in AI adoption
According to the results, companies consider the shortage of skilled workers in the AI sector and a lack of understanding of the technology to be key challenges. Regulatory requirements - particularly in terms of data protection - would make the introduction of artificial intelligence even more difficult. Many SMEs want transparent and reliable solutions as they want to validate AI outputs and clarify liability issues. Generative AI services such as ChatGPT are constantly evolving, but according to the study results, companies often lack customized solutions based on self-trained models and their own data.
"The technology is making progress, but without targeted investment in employee training and the creation of transparent processes, sustainable adoption cannot take place. Human and organizational factors are crucial to fully exploiting the potential of AI," says Evangelos Xevelonakis, Head of the Center for Data Science & Technology at the HWZ.
"The results show that Swiss companies are very interested in data-driven management, but that the development of specialized AI solutions is often hindered by a lack of expertise and uncertainty about the ROI," adds Matthias Mohler, Head of Data & AI Consulting at Swisscom.
Chandler Robinson, a student of the MAS Applied Data Science at the HWZ, conducted the study as part of his Master's thesis. According to the report, he received support from Evangelos Xevelonakis from the HWZ and Matthias Mohler from Swisscom. The survey was conducted in fall 2024 and provides insights into the use of AI by 123 Swiss SMEs and 5 large companies. In addition to the survey, Robinson conducted seven personal interviews with senior IT executives. (Dylan Wildhaber/jor/swi)
