University of Illinois Urbana-Champaign

AI and Human Values in Homelessness Services: How Values Can Inform the Design of AI for Social Services

Suzuki, Takayuki; Fleischmann, Kenneth R.; Slota, Stephen C.; Smarda, Daniel; Morris-King, James R.; Gertner, Abigail S.

Permalink

https://hdl.handle.net/2142/122785

Description

Title
AI and Human Values in Homelessness Services: How Values Can Inform the Design of AI for Social Services
Author(s)
  • Suzuki, Takayuki
  • Fleischmann, Kenneth R.
  • Slota, Stephen C.
  • Smarda, Daniel
  • Morris-King, James R.
  • Gertner, Abigail S.
Issue Date
2024-03-20
Keyword(s)
  • Artificial intelligence
  • Human values
  • Homelessness
  • Value-sensitive design
Abstract
The goal of this paper is to understand how social service providers’ work values shape their views on how artificial intelligence (AI) can be used to improve social service provision for people experiencing homelessness. The paper builds on work in value-sensitive design and the use of AI for social service provision. This paper reports findings from surveys collected from 207 representatives of relevant programs funded by the U.S. Department of Housing and Urban Development. Participants were asked to complete two value instruments – one covering their own values in relation to homeless service provision work (work values) and one covering values in relation to AI (AI values). The survey responses yielded four clusters of work values (community, security, effectiveness, and inclusivity) and three clusters of AI values (properties of AI, human-AI interaction, and broader impacts of AI). The results identified relationships between the work clusters and the AI clusters; for example, community was positively correlated with human interaction with AI (p<0.01) and broader impacts of AI (p<0.001); effectiveness was positively correlated with properties of AI (p<0.01); and inclusivity was positively correlated with broader impacts of AI (p<0.001). These results demonstrate that knowledge of stakeholders’ values toward homeless service provision can be used to inform the design of AI-based systems for social service provision.
Publisher
iSchools
Series/Report Name or Number
iConference 2024 Proceedings
Type of Resource
Other
Language
eng
Handle URL
https://hdl.handle.net/2142/122785
Copyright and License Information
Copyright 2024 is held by Takayuki Suzuki, Kenneth R. Fleischmann, Stephen C. Slota, Daniel Smarda, James R. Morris-King and Abigail S. Gertner. Copyright permissions, when appropriate, must be obtained directly from the authors.

Owning Collections

iConference 2024 Posters PRIMARY
Posters presented at the 2024 iConference https://www.ischools.org/iconference