In December, Research ICT Africa (RIA) submitted a comprehensive response to the UN Secretary General’s policy brief about Information Integrity on Digital Platforms. The policy brief offers principles for a UN Code of Conduct, which UN Member States and digital media companies would be asked to follow.
RIA’s response, Input for: Our Common Agenda Policy Brief 8 – Information Integrity on Digital Platforms, reflects our work in this area, including research for UNESCO, which extensively assesses the nexus of platforms, disinformation, and researcher access to data.
The points made in the RIA submission:
- The Code could include provisions to tackle ethical and self-regulatory issues for the commercial producers of disinformation. The UNSG’s document could add: “Public relations and advertising groups and their associations should adhere to codes of conduct for truth in advertising, and act against members who sell services which promote disinformation and hate speech.”
- There could be explicit reference to the importance of human rights impact assessments (HRIAs). Such tools help to anticipate and mitigate threats to information integrity.
- The Code recognises that independent media is an alternative to disinformation. RIA urges stronger wording here, like: “Governments, donors, and foundations should be urged to increase support to the viability of independent media and to the exercise of investigative journalism, including by backing the International Fund for Public Interest Media.” (for which the Secretary-General has expressed support). Further, RIA recommended adding: “Digital platforms should recognise credible news outlets and adapt recommender systems to uprank their content”.
- If the Code proposed that Member States make platforms legally liable to be transparent, says the RIA submission, these companies will be incentivised to increase their actions to anticipate and mitigate content risks to human rights.
- For the Code to do justice to user empowerment, it should urge that stakeholders can get access not just to information but also to data, including both public and private data.
- RIA welcomes the UN Secretary General’s point that “data access for researchers is also an urgent priority on a global scale”. The submission urges that the text goes further to encourage States to “support independent mechanisms with appropriate safeguards and vetting procedures”, and that efforts are also encouraged to assist researchers with appropriate tools and digital research capacity.
- To disincentivise the production and circulation of disinformation, the RIA submission proposes that the Code calls for credible verification steps for platforms allocating “blue ticks” which imply authentic online identity. RIA also highlights the importance of watermarking and labelling that shows content provenance.
- The submission suggests that the Code’s reference to safety in AI, should cover both foundation models and applications in the field of generative AI, as well as the classificatory AI deployments that are used in recommender systems and advertising technology.
- Also advised is that the Code propose to UN Member States to support platform pluralism, including support for decentralised and non-profit alternatives to the major companies. This would provide more choice to users, particularly for platforms in local languages and at a scale where safety is manageable.
Finally, the RIA submission urges the UN and its agencies to support multistakeholder panels to conduct assessments in given countries, which in turn can produce reports on shortfalls and successes in relation to the Code.
