The use of Large Language Models (LLMs) to counter problematic smartphone use and support users’ digital wellbeing has recently gained research interest. Yet, such an approach is still in its infancy, particularly when compared to traditional digital self-control interventions. In this paper, we explore the possibility of using LLMs as “digital wellbeing assistants.” Specifically, we first reviewed the HCI literature and developed four user personas that exemplify widely recognized issues associated with smartphone (over)use. Then, we assessed the capabilities of four popular LLMs-powered chatbots, i.e., Bing, ChatGPT, Gemini, and Claude.AI, in understanding problematic smartphone uses and suggesting practical strategies to address them, using the developed personas as a testing ground. Despite some variations, results show that all three LLMs can offer tailored suggestions based on user characteristics, opening doors for smarter digital self-control interventions that leverage AI to support users’ self-monitoring and regulation capabilities.

Dialogues with digital wisdom: can LLMs help us put down the phone? / De Russis, Luigi; Monge Roffarello, Alberto; Scibetta, Luca. - ELETTRONICO. - (2024), pp. 56-61. (Intervento presentato al convegno ACM 4th International Conference on Information Technology for Social Good (GoodIT 2024) tenutosi a Bremen (DEU) nel 04 September 2024) [10.1145/3677525.3678640].

Dialogues with digital wisdom: can LLMs help us put down the phone?

De Russis, Luigi;Monge Roffarello, Alberto;Scibetta, Luca
2024

Abstract

The use of Large Language Models (LLMs) to counter problematic smartphone use and support users’ digital wellbeing has recently gained research interest. Yet, such an approach is still in its infancy, particularly when compared to traditional digital self-control interventions. In this paper, we explore the possibility of using LLMs as “digital wellbeing assistants.” Specifically, we first reviewed the HCI literature and developed four user personas that exemplify widely recognized issues associated with smartphone (over)use. Then, we assessed the capabilities of four popular LLMs-powered chatbots, i.e., Bing, ChatGPT, Gemini, and Claude.AI, in understanding problematic smartphone uses and suggesting practical strategies to address them, using the developed personas as a testing ground. Despite some variations, results show that all three LLMs can offer tailored suggestions based on user characteristics, opening doors for smarter digital self-control interventions that leverage AI to support users’ self-monitoring and regulation capabilities.
2024
9798400710940
File in questo prodotto:
File Dimensione Formato  
main.pdf

accesso aperto

Tipologia: 2. Post-print / Author's Accepted Manuscript
Licenza: PUBBLICO - Tutti i diritti riservati
Dimensione 814.28 kB
Formato Adobe PDF
814.28 kB Adobe PDF Visualizza/Apri
Pubblicazioni consigliate

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11583/2992025