It feels like AI is everywhere at the moment – or will be quite soon. The media is no exception. But while the technology is developing rapidly in the global north, much of the global south is having to play catch up. We wrote about this recently.
A speaker at the Media Freedom Lab said the information AI is fed is 80% in English and based on Caucasian experiences and standards, creating a linguistic and racial bias within the data. We hosted this event to address how to decolonise AI and why this is important.
One of the biggest take-away points was that whilst there are undoubtedly advantages for journalists and filmmakers, the divide between how AI was being developed, used and received in the global south and global north risks creating cultural apartheid. There was consensus that more ethical practices need to be considered when thinking about the impact of AI in different countries.
Our Media Freedom Lab, which was made up of two panel discussions and lightning presentations, provided an opportunity for attendees to engage in discussions on how the rapid technological change was being experienced by different groups and the innovative use of AI in documentaries and projects.
5 key ideas for building more ethical principles into AI
These points reflect a summary of the discussion. We deliberately have not attributed the comments to allow the contributors to freely express themselves.
There needs to be more equity of resourcing
Media organisations and newsrooms based in the global south lack the infrastructure to develop AI with the same intensity as the global north, which means they risk not being able to benefit from its positive attributes. For many, AI is not seen as a priority as they have more fundamental problems to deal with like being able to operate safely and freely. More investment needs to be made available, so they are not left behind, and have the agency to build AI using their own vision.
Collaboration between multiple stakeholders
The data used in AI needs to include a multiplicity of voices and experiences, considering language variations within countries. Developers need to make more of a conscious effort to address this by engaging with wider audiences, including grassroots organisations, indigenous communities and civic society.
Education is key
You can’t use a new technology if you don’t know what it’s for or what its limitations are. We must educate a new generation to understand AI in a responsible way and map out its capabilities. At the same time, we must not ignore older generations as there is a need to build an understanding of AI for all parts of society.
Regulation with caution
Media regulators need to be more informed about AI if they can ensure responsible practices are built into its rapid growth, particularly with regards to copyright and licensing, and the impact on the economic sustainability of newsrooms and journalists. They need to engage more with journalists and tech developers of all sizes to ensure regulation is fair and workable. However, increasing media regulatory powers in some countries could be used to curtail media freedom and independent journalism. One idea was to develop a new legal framework just for AI.
Critical thinking on AI
Our final take-away point is that no one should be afraid of voicing scepticism about AI. We should always question the bias in tech, be transparent about its harmful impact on the environment, and monitor where its growth relies on the use of indentured labour. We need to hold the powerful accountable.
OWM looks forward to continuing this conversation and building collaborations.





