Unlawful by design: Exposing the human rights costs of generative AI (www.amnesty.org)

A recent briefing by Amnesty International highlights significant human rights concerns surrounding standalone generative AI systems, which utilize unlawful web scraping for data collection. The report argues that these technologies not only breach international human rights law but also perpetuate privacy violations, discrimination, and threats to freedom of expression. Amnesty’s findings indicate that the foundational practices behind these AI systems—rooted in mass data invasions—render them fundamentally incompatible with human rights standards. The significance of this report for the AI/ML community lies in its call for a prohibition on such generative AI systems, urging developers and policymakers to rethink how these technologies are designed and implemented. By exposing the ethical and legal implications of current data practices, Amnesty International advocates for a shift toward more responsible AI development that respects privacy rights and complies with international norms. This briefing serves as a critical reminder of the potential societal costs associated with unchecked AI advancements, pushing for an urgent reevaluation of AI deployment strategies.