S.Africa AI Association on responsible AI in social media

The South African Artificial Intelligence Association has recently raised concerns about LinkedIn's AI model training practices, accusing the platform of breaching aspects of the Protection of Personal Information Act (POPIA). Nathan-Ross Adams, Advisory Board Member at the South African Artificial Intelligence Association, shed light on the issue in a recent interview on CNBC Africa. Adams explained that the association is committed to promoting trustworthy AI for all stakeholders and believes in the importance of ethical AI practices. The controversy arose when Adams discovered that LinkedIn had automatically opted-in its users to allow the platform to train its AI models using their personal data. LinkedIn, a popular professional networking site, boasts a significant user base in South Africa, with 12 million users as of a recent data report in 2024. By enabling AI model training based on user data, LinkedIn could potentially utilize this information for various purposes such as targeted advertising, gaining behavioral insights, and even identifying areas for new business ventures requiring specific expertise found on the platform. The AI Association deemed this practice as a violation of the POPIA regulations and raised formal complaints with the relevant authorities. Adams highlighted the key issue of consent under POPIA, emphasizing that consent for processing personal data must be specific, informed, and voluntary. By automatically opting-in South African users without providing clear information on the implications of such data processing, LinkedIn failed to meet the standards set by the legislation. As a result, the AI Association urged the regulators to take action against LinkedIn to ensure compliance with data protection laws. In response to the controversy, LinkedIn users also expressed concerns about the transparency and accountability of other tech platforms regarding their AI practices. While acknowledging the potential benefits of AI technology, users emphasized the importance of trust and ethical considerations in AI development. Adams stressed the need for greater transparency from all tech companies, emphasizing the fundamental role of trust in the future co-existence with AI. Looking ahead, the South African government faces the challenge of regulating AI technologies effectively to protect citizens' data rights and ensure ethical AI practices. The global AI race between Western and Eastern powers has intensified, raising questions about the influence of advanced AI on societal values and principles. As South Africa aligns with Western democracies, the government must prioritize understanding and regulating AI technologies to safeguard its citizens' interests. The AI Association's advocacy for responsible AI practices and data protection serves as a crucial reminder of the ethical considerations that must underpin technological advancements. As the digital landscape evolves, stakeholders must collaborate to establish robust regulatory frameworks and guidelines to uphold ethical standards in AI development and deployment.