Filter by author: Markus Lehmkuhl

Engaging with the ongoing debate regarding the portrayal of artificial intelligence (AI) in the public sphere – particularly the alleged predominance of sci-fi imagery and humanoid robots – our study examines how six German print media visualize articles related to AI. A mixed-methods approach combines qualitative and quantitative visual content analysis, analyzing 818 images from articles published in 2019 and 2022/23. Our findings indicate that human figures, rather than robots, serve as dominant visual objects, and no pronounced gaps between textual and visual representations of AI were observed. Overall, German print media appear to present a differentiated perspective on AI, balancing opportunities and risks associated with this technology.

We assessed ChatGPT's ability to identify and categorize actors in German news media articles into societal groups. Through three experiments, we evaluated various models and prompting strategies. In experiment 1, we found that providing ChatGPT with codebooks designed for manual content analysis was insufficient. However, combining Named Entity Recognition with an optimized prompt for actor Classification (NERC pipeline) yielded acceptable results. In experiment 2, we compared the performance of gpt-3.5-turbo, gpt-4o, and gpt-4-turbo, with the latter performing best, though challenges remained in classifying nuanced actor categories. In experiment 3, we demonstrated that repeating the classification with the same model produced highly reliable results, even across different release versions.