April 7, 2024

This Week in Responsible AI: Apr 7, 2024

This Week in Responsible AI: Apr 7, 2024

Bias

• Meta’s AI image generator can’t imagine an Asian man with a white woman

• Measuring Political Bias in Large Language Models: What Is Said and How It Is Said

• Survey of Bias In Text-to-Image Generation: Definition, Evaluation, and Mitigation

Law/Policy

• Automated Transparency: A Legal and Empirical Analysis of the Digital Services Act Transparency Database

• Washington state judge blocks use of AI-enhanced video as evidence in possible first-of-its-kind ruling

AI and Knowledge

• AI and the Problem of Knowledge Collapse

• LLMs meet misinformation

• China tests US voter fault lines and ramps AI content to boost its geopolitical interests

Jailbreaking LLMs

• Jailbreaking Leading Safety-Aligned LLMs with Simple Adaptive Attacks

• Many-shot jailbreaking

• JailbreakBench: An Open Robustness Benchmark for Jailbreaking Large Language Models

Data

• For AI firms, anything "public" is fair game

• AI, Encryption, and the Sins of the 90s (NDSS 2024 Keynote by Meredith Whittaker)

Other

• NLP for Maternal Healthcare: Perspectives and Guiding Principles in the Age of LLMs

• Resistance in the Black Box Society

• ‘Lavender’: The AI machine directing Israel’s bombing spree in Gaza

• The mechanisms of AI hype and its planetary and social costs

• Meta plans to more broadly label AI-generated content

• Context Before Code: Meta’s Oversight Board Policy Advisory Opinion on the Word “Shaheed” Calls for Language and Cultural Nuance in Content Moderation

Compiled by Leif Hancox-Li