Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Sociotechnical Implications of Generative Artificial Intelligence for Information Access

Bhaskar Mitra, Henriette Cramer, Olya Gurevich

TL;DR

This chapter analyzes the sociotechnical implications of generative AI for information access, arguing that robust information access requires attention to systemic consequences beyond content quality. It adopts the Consequences-Mechanisms-Risks (CMR) framework to map five negative outcomes—information ecosystem disruption, concentration of power, marginalization, innovation decay, and ecological impact—onto concrete mechanisms and societal risks, drawing on a wide literature base. It then proposes a comprehensive evaluation and mitigation toolkit spanning threat modeling, lifecycle risk assessment, benchmark-context alignment, RAG-based factuality checks, and safe-release practices, while advocating for open sharing of evaluation methods. The piece also surveys incentives and organizational dynamics, highlighting data labor, industry capture, and governance structures, and calls for a whole-of-society approach to align AI deployment with democratic values, health literacy, and climate sustainability. Collectively, it offers a structured framework to guide researchers and policymakers toward safer, more equitable information access systems in the era of generative AI, emphasizing civil oversight and pluralist research agendas.

Abstract

Robust access to trustworthy information is a critical need for society with implications for knowledge production, public health education, and promoting informed citizenry in democratic societies. Generative AI technologies may enable new ways to access information and improve effectiveness of existing information retrieval systems but we are only starting to understand and grapple with their long-term social implications. In this chapter, we present an overview of some of the systemic consequences and risks of employing generative AI in the context of information access. We also provide recommendations for evaluation and mitigation, and discuss challenges for future research.

Sociotechnical Implications of Generative Artificial Intelligence for Information Access

TL;DR

This chapter analyzes the sociotechnical implications of generative AI for information access, arguing that robust information access requires attention to systemic consequences beyond content quality. It adopts the Consequences-Mechanisms-Risks (CMR) framework to map five negative outcomes—information ecosystem disruption, concentration of power, marginalization, innovation decay, and ecological impact—onto concrete mechanisms and societal risks, drawing on a wide literature base. It then proposes a comprehensive evaluation and mitigation toolkit spanning threat modeling, lifecycle risk assessment, benchmark-context alignment, RAG-based factuality checks, and safe-release practices, while advocating for open sharing of evaluation methods. The piece also surveys incentives and organizational dynamics, highlighting data labor, industry capture, and governance structures, and calls for a whole-of-society approach to align AI deployment with democratic values, health literacy, and climate sustainability. Collectively, it offers a structured framework to guide researchers and policymakers toward safer, more equitable information access systems in the era of generative AI, emphasizing civil oversight and pluralist research agendas.

Abstract

Robust access to trustworthy information is a critical need for society with implications for knowledge production, public health education, and promoting informed citizenry in democratic societies. Generative AI technologies may enable new ways to access information and improve effectiveness of existing information retrieval systems but we are only starting to understand and grapple with their long-term social implications. In this chapter, we present an overview of some of the systemic consequences and risks of employing generative AI in the context of information access. We also provide recommendations for evaluation and mitigation, and discuss challenges for future research.
Paper Structure (30 sections, 2 figures, 2 tables)

This paper contains 30 sections, 2 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Implications of generative AI for information access
  3. Consequences and mechanisms
  4. Consequence: Information ecosystem disruption
  5. Consequence: Concentration of power
  6. Consequence: Marginalization
  7. Consequence: Innovation decay
  8. Consequence: Ecological impact
  9. Risks
  10. Risks to society
  11. Risks to IR research
  12. Risks to environment
  13. Methods to evaluate risks and impact
  14. Evaluating the impact of generative IR applications
  15. Threat identification, assessment, and modeling
  16. ...and 15 more sections

Figures (2)

  • Figure 1: Primary actors responsible for aligning technology with societal interests
  • Figure 2: Mitra's mitra2024search hierarchy of IR stakeholder needs. More critical needs are at the bottom of the pyramid. This figure has been reproduced from the original paper with permission.