[Paper Review] Anticipating Impacts: Using Large-Scale Scenario Writing to Explore Diverse Implications of Generative AI in the News Environment

Kimon Kieslich, Nicholas Diakopoulos+1 more|arXiv (Cornell University)|Oct 10, 2023

Misinformation and Its Impacts4 citations

TL;DR

This study uses large-scale scenario writing and participatory foresight with 119 stakeholders—news consumers, developers, and content creators—to explore diverse anticipated impacts of generative AI in the news environment. It identifies key risks like disinformation and bias, maps mitigation strategies, and reveals significant differences in stakeholder perceptions, particularly regarding transparency obligations under the EU AI Act, demonstrating scenario writing as a valuable tool for AI impact assessment.

ABSTRACT

The tremendous rise of generative AI has reached every part of society - including the news environment. There are many concerns about the individual and societal impact of the increasing use of generative AI, including issues such as disinformation and misinformation, discrimination, and the promotion of social tensions. However, research on anticipating the impact of generative AI is still in its infancy and mostly limited to the views of technology developers and/or researchers. In this paper, we aim to broaden the perspective and capture the expectations of three stakeholder groups (news consumers; technology developers; content creators) about the potential negative impacts of generative AI, as well as mitigation strategies to address these. Methodologically, we apply scenario writing and use participatory foresight in the context of a survey (n=119) to delve into cognitively diverse imaginations of the future. We qualitatively analyze the scenarios using thematic analysis to systematically map potential impacts of generative AI on the news environment, potential mitigation strategies, and the role of stakeholders in causing and mitigating these impacts. In addition, we measure respondents' opinions on a specific mitigation strategy, namely transparency obligations as suggested in Article 52 of the draft EU AI Act. We compare the results across different stakeholder groups and elaborate on the (non-) presence of different expected impacts across these groups. We conclude by discussing the usefulness of scenario-writing and participatory foresight as a toolbox for generative AI impact assessment.

Motivation & Objective

• To broaden impact assessment of generative AI in news beyond developer and researcher perspectives by including news consumers and content creators.
• To identify and categorize anticipated negative impacts of generative AI on the news environment from diverse stakeholder viewpoints.
• To map stakeholder-specific mitigation strategies for generative AI risks in journalism.
• To evaluate stakeholder attitudes toward transparency obligations, particularly Article 52 of the draft EU AI Act.
• To assess the utility of scenario writing and participatory foresight as tools for anticipatory governance of generative AI.

Proposed method

• Conducted a survey with 119 participants across three stakeholder groups: news consumers, technology developers, and content creators.
• Employed participatory foresight through structured scenario-writing exercises to elicit cognitively diverse visions of future AI impacts in news.
• Used thematic analysis to systematically code and categorize scenarios into themes of impact, mitigation, and stakeholder roles.
• Measured opinions on transparency obligations as proposed in Article 52 of the draft EU AI Act to assess stakeholder support.
• Compared stakeholder group responses to identify differences in perceived risks and preferred solutions.
• Integrated qualitative findings with quantitative opinion data to assess consensus and divergence across groups.

Experimental results

Research questions

• RQ1 What types of negative impacts do different stakeholder groups anticipate for generative AI in the news environment?
• RQ2 How do stakeholder groups differ in their perceptions of the causes and solutions for generative AI-related risks in journalism?
• RQ3 What role do transparency obligations, such as those in Article 52 of the EU AI Act, play in stakeholder expectations for mitigating AI risks?
• RQ4 What mitigation strategies do stakeholders propose for issues like disinformation, bias, and social polarization in AI-driven news?
• RQ5 How effective is scenario writing and participatory foresight in capturing diverse, anticipatory perspectives on generative AI impacts?

Key findings

• News consumers most frequently anticipated generative AI’s role in spreading disinformation and undermining trust in news media.
• Technology developers were more likely to emphasize technical risks such as model bias and data quality issues, while also expressing cautious optimism about AI’s potential to improve news production.
• Content creators highlighted concerns about job displacement and the erosion of editorial standards due to AI-generated content.
• Support for transparency obligations under Article 52 of the EU AI Act was strongest among news consumers and content creators, with lower support among developers.
• Thematic analysis revealed distinct clusters of impact—such as credibility erosion, algorithmic amplification, and labor market shifts—each with unique mitigation strategies proposed by different stakeholder groups.
• The study demonstrates that scenario writing effectively captures diverse, context-specific expectations of AI impacts, offering a robust method for anticipatory governance.