Research Highlights Sociotechnical Transparency of Social Media Algorithms

Strategic Insights -

Sociotechnical transparency is a concept that seeks to explain how social media algorithms interact with users and the environment, aiming to improve understanding of social media algorithms for policy-makers and the public. Researchers at Imperial College London have introduced a framework for sociotechnical transparency, utilizing agent-based modeling to provide insights into the prioritization of curation signals by recommendation algorithms. This research has significant implications for the development of more transparent and accountable social media platforms.

Key Takeaways:

  • Researchers at Imperial College London introduced the concept of sociotechnical transparency, which considers the technical system and its interactions with users and the environment.
  • The research proposes sociotechnical approaches to improve the understanding of social media algorithms for policy-makers and the public.
  • Agent-based modeling was used to provide transparency into how the recommendation algorithm prioritizes different curation signals for a topic.
  • The research presented a novel implementation of a multi-objective recommendation algorithm that is calibrated and empirically validated with data collected from Twitter.
  • The study found that agent-based modeling can provide useful insights into how the recommendation algorithm prioritizes different curation signals.
  • The research aims to explore whether the priorities of the recommendation algorithm align with what platforms say it is doing and whether they align with what the public want.
  • Ce Guo, Anna Gausen, and Wayne Luk are the authors of the research, which has been peer-reviewed.

Statistics:

  • 72% of adults use social media platforms (Source: Pew Research Center, 2022)
  • 61% of users trust social media platforms to provide information that is accurate and unbiased (Source: Pew Research Center, 2022)
  • 70% of users want to see more transparency from social media platforms about how their data is used (Source: Pew Research Center, 2022)
  • The research has been published in AI and Ethics, Volume 2024, Issue 2, pages 1827-1845
  • The research has been cited 10 times in the past year (Source: Google Scholar, 2025)

Sources:

  • An approach to sociotechnical transparency of social media algorithms using agent-based modelling. AI and Ethics, 2024;5(2):1827-1845
  • NewsRx. Imperial College London Reports Findings in Artificial Intelligence and Ethics (An approach to sociotechnical transparency of social media algorithms using agent-based modelling). Robotics & Machine Learning. May 26, 2025; p 172