AI BriefingAmnesty International Warns of Human Rights Violations by Generative AI, Calls for Ban
Key point
Amnesty International claimed that major generative AI models violate international human rights law and urged government regulation.
Details
Amnesty International released a report titled 'Unlawful by Design: Exposing the Human Rights Costs of Generative AI', declaring that major generative AI products including GPT-3, Gemini, Llama, DeepSeek, Midjourney, and Stable Diffusion are fundamentally incompatible with international human rights law.
The key points of the report are as follows:
- Privacy Violations: Collecting training data through large-scale web scraping constitutes a mass privacy violation without the knowledge or consent of the individuals concerned.
- Discrimination and Bias: Racial and gender biases embedded in training data perpetuate discrimination.
- Freedom of Expression: AI moderation systems have been found to excessively censor non-English content.
- Psychological Manipulation: Manipulating users' intentions and thought processes through predictive suggestions may constitute coercion rather than a mere design flaw.
Amnesty International requested comments from major companies including Google, OpenAI, Meta, Microsoft, and Amazon regarding this matter, but only 5 companies responded.
This summary was generated automatically by AI. Check the original for the author's claims and context. Copyright belongs to the original author.
Our guide explains how the AI works. Report summary errors, attribution issues, or removal requests via Contact.
Related
Class Action Suit Alleges OpenAI Routed ChatGPT Conversations to Human Reviewers Without Clear Disclosure
NSA Spending Billions to Test Frontier AI Models for Security Vulnerabilities
Comfy Launches Comfy Router: A Unified API for Frontier Media Models