MISA urges human rights- centric AI regulation amid growing concerns for free expression in Africa

By Dumisani Ndlovu

HARARE– The Media Institute of Southern Africa (MISA) has submitted comprehensive insights to the African Commission on Human and Peoples’ Rights (ACHPR), advocating for the establishment of Artificial Intelligence (AI) legal frameworks firmly rooted in international human rights law.
The submission, made electronically on July 23, 2025, indicate critical concerns about AI's potential impact on freedom of expression, journalism, and the information landscape across Africa.
As a prominent advocate for media freedom and digital rights, MISA stressed the necessity for governments to ensure AI development incorporates transparency, accountability, data security, and robust redress mechanisms. The organization emphasized that to safeguard fundamental human rights, essential safeguards, including mandatory human oversight, must be integrated throughout the entire lifecycle of any AI system impacting these rights.
MISA highlighted several pressing concerns, noting how Generative Artificial Intelligence (GenAI) has amplified misinformation, blurring the lines between truth and fiction and posing a risk to journalistic editorial independence. The rise of deepfakes, capable of political manipulation, character assassination, or inciting violence, was also identified as a significant threat.
The submission pointed out that most AI systems are trained on predominantly Western datasets, leading to inherent biases that can perpetuate discrimination against African populations and misrepresent local contexts. Furthermore, MISA warned against the potential for state control and censorship, citing the increased use of AI for surveillance, such as facial recognition and social media monitoring, which can lead to self-censorship among journalists and citizens.
The dominance of big tech companies, controlling much of the AI landscape, was identified as a threat to smaller media outlets, with the monetization of data frequently reflecting commercial biases and creating economic dependencies that undermine editorial independence. Compounding these issues, MISA also underscored the persistent digital divide and resource gaps in many African countries, where a lack of affordable and reliable internet hinders citizens from fully leveraging emerging technologies. The organization also noted that many international AI instruments are non-binding and often lack perspectives from the Global South, creating challenges in practical policy implementation.
In the future , MISA prescribed a multi-pronged approach as it reiterated the call for governments to establish AI legal frameworks grounded in international human rights law, ensuring transparency, accountability, data security, and clear redress mechanisms.
The organization also recommended that the media industry develop its own AI Policy or Code of Ethics, mandating clear labeling for all AI-generated or altered content and ensuring rigorous human review and editorial approval before publication, particularly for news and informational purposes. Crucially, AI systems must not be allowed to influence editorial independence or journalistic decisions.
MISA also stressed the importance of designing AI systems inclusively, incorporating input from diverse stakeholders, including marginalized groups and underrepresented voices. To address the digital divide, strong policies are needed to ensure affordable and accessible internet connectivity, alongside enhanced digital literacy initiatives for communities.
The organization called for greater accountability in the utilization of Universal Service Funds to stimulate infrastructure development, bridging the gap between urban and rural areas. Finally, MISA emphasized that regional and global coordination is vital to harmonize AI development and translate principles into practical, enforceable policies through multi-stakeholder partnerships.