Sense about Science—written evidence (FON0042)

House of Lords Communications and Digital Select Committee inquiry: The future of news: impartiality, trust, and technology

Sense about Science

Sense about Science is an independent charity that promotes the public interest in sound science and evidence. We work with communities all over the UK to make sense of evidence, and with researchers and policy makers to raise the standard of evidence in public life. We have 20 years of experience working with experts and the public to create our making sense guides that provide clear and balanced evidence behind important and often misunderstood areas of science.

1. What impacts (positive and negative) do large technology platforms and online news aggregators have on the UK’s news environment, including media plurality? And how might this change?

Algorithms are deciding more and more what content users can see and engage with. The lack of transparency in how these algorithms operate presents a significant risk of people being shown biased and partial information.

Negligent curation of information

Content is surfaced by platforms using algorithms designed to serve the attention economy – their objective is to encourage user engagement, not provide the optimal information. When identical searches by different users produce very different results, it is possible for people to consistently receive weak or biased information that does not help them develop a balanced understanding. While each item surfaced might individually not meet the algorithm’s harm threshold, they cumulatively provide a skewed picture of the world, with users invited to seek out more extreme content in the algorithm’s push for engagement.

When the main purpose of technology platforms is to keep users online, they act as ‘negligent curators’ choosing what to surface for their own ends rather than that of users, suggesting who to follow and what news to read.

Censorship of good information

Platforms have begun deploying algorithms to moderate online content in order to identify and remove prohibited content. Currently these algorithms are opaque, with no external oversight or transparency of decision-making processes, which risks moderation turning into a crude censorship.

We have already seen the removal or demotion of good quality information and important discussions of difficult issues or conflicting evidence, as a result of arbitrary censorship. For example, Facebook tagging a credible BMJ article as ‘partly false’ because 'the authors did not express unreserved support for vaccination’, or Instagram ‘shadow banning’ a Cochrane (Harding Prize for Good Science Communication winning) review for ‘false content’, which was also tagged by Twitter as ‘misleading’.[1]

Science research and development is particularly vulnerable to erroneous moderation by AI systems, since these learn on past data and established consensus, and are biased against new and counterfactual content regardless of its integrity. Moderation systems must also be able to identify and interpret contextual and social nuances of the human language.

5. Are there any actions the Government should take to address concerns around due impartiality, trust, and the influence of technology platforms?

a) Are changes needed to the Media Bill?

Censorship is not a sustainable long-term solution to address the growing challenges from mis- and disinformation. People have become accustomed to blaming other users for putting up poorly judged or ill-informed content, but the real question should be why the content has made it into people’s feed. The lack of transparency over the algorithm software has meant it’s near impossible to understand what criteria information is being curated under or why certain information is being banned. Commercial confidentiality is not a good enough reason to ignore the risks from using unaccounted algorithms to decide what content is appropriate. This risks people not engaging in difficult issues and not being able to have honest discussions where uncertainty and a lack of evidence exist.

Without knowing the data used to build the algorithm and the underlying assumptions used to construct it, we can't be sure if it is able to distinguish between harmful and benign content more effectively than a human moderator. To ensure the algorithm is working as intended, the processes by which information is shared/promoted or suppressed/censored should be open to public scrutiny.

b) Are changes needed to the way the Government addresses mis- and dis-information?

The best way to counter mis- and disinformation is to provide good access to useful information. The lack of functioning regulation of online content, resulting from arbitrary censorship, means people online are faced with false information or information of variable quality. Yet the public are trying to understand various issues where the information is either incomplete, unreliable or placed out of context. By failing to share or promote good information the government is leaving people who wish to counter misinformation themselves with nothing they can point to reliably as a source of information. People cannot prevent ‘false narratives’ without providing a full account of the true narrative.

Rather than facing new limits on forms of communication or blame individuals for sharing mis- and disinformation, people need to be equipped to find good information and work out how reliable it is. Restricting access to good information should therefore be recognised as equally harmful as promoting misinformation.

12 February 2024

3