Abstract ID: 3.11117 | Accepted as Poster | Talk/Oral | TBA | TBA

Precarious livelihoods in mountain regions (Wilson & Dashper, 2023) fueled among other reasons by the cost of living crisis (Cró & Martins, 2024) as well as the tarnished employer image (Mölk et al., 2022) pose an extraordinary challenge for tourism employers that aim to hire and retain workforce (Liu-Lastres et al., 2023). Furthermore, employees may also face psychological pressure due to the enhanced importance of generative artificial intelligence (AI) in the workplace (Liu et al., 2024). Thus, it is crucial to understand: How do employees and job seekers perceive employers and working conditions in tourism during the generative AI shift?

The human researcher will collaborate with an large-language-model (LLM) (Arora et al., 2025) for the purpose of a literature review (Bichler et al., 2022). While the human author will define keywords based on crucial papers (Liu-Lastres et al., 2024; Mölk et al., 2022) in order to derive literature from the Web-of-Science (Newbert, 2007), the LLM will be an aide during the review process of the abstracts. Via Retrieval-Augmented-Generation (RAG) abstracts found in the Web-Of-Science database will be provided to an LLM (Arora et al., 2025). The LLM can obtain the abstracts and query the texts via a chain-of-thought (CoT) process (Arora et al., 2025; Wei et al., 2022). The results can validate the human assessment of the researcher, who also will take the abstracts into consideration (Sarkis-Onofre et al., 2021). Therefore, the literature is approached in a way that combines strengths from the PRISMA method and contemporary LLMs.