March 3, 2024

This Week in Responsible AI: Mar 3, 2024

This Week in Responsible AI: Mar 3, 2024

Faces

• Face off: Attackers are stealing biometrics to access victims’ bank accounts

• Vending machine error reveals secret face image database of college students

Copying

• How artists are fighting generative AI

• Copyleaks Research Finds Nearly 60% of GPT-3.5 Outputs Contained Some Form of Plagiarized Content

Participation

• Wikibench: Community-Driven Data Curation for AI Evaluation on Wikipedia

• Controversies, contradiction, and “participation” in AI

• Meaningful public participation and AI

Gemini

• Ethical AI Isn’t to Blame for Google’s Gemini Debacle

• Google's Gemini refuses to say whether Elon Musk tweeting memes or Hitler negatively impacted society more

• Google pauses Gemini’s ability to generate AI images of people after diversity errors

Law/Policy

• Regulating for trust: Can law establish trust in artificial intelligence?

• Utah's AI law

• Price fixing by algorithm is still price fixing

• Unregulated AI could cause the next Horizon scandal

• Amba Kak creates policy recommendations to address AI concerns

• Between Copyright and Computer Science: The Law and Ethics of Generative AI

• What the US AI Safety Consortium means for enterprises

• OpenAI says New York Times 'hacked' ChatGPT to build copyright lawsuit

• ICO orders Serco Leisure to stop using facial recognition technology to monitor attendance of leisure centre employees

Transparency

• Austria’s social security invests over €50m in AI – just for bookkeeping?

• Apple Investors Reject Call for Report Into Company’s AI Use

• Fighting in the dark: How Europeans push back against rogue AI

Fakes

• Seeking Reliable Election Information? Don’t Trust AI

• Behind a Secretive Global Network of Non-Consensual Deepfake Pornography

• Here lies the internet, murdered by generative AI

• A Beverly Hills middle school is investigating students sharing AI-made nude photos of classmates

• Seeking Reliable Election Information? Don’t Trust AI

• Hallucination is Inevitable: An Innate Limitation of Large Language Models

• Piers Morgan and Oprah Winfrey 'deepfaked' for US influencer's ads

• Are outright bans on deepfakes feasible?

Other

• Gemma's Responsible Generative AI Toolkit

• Accelerating Progress Toward Trustworthy AI

• Arbitrariness and Prediction: The Confounding Role of Variance in Fair Classification

• Hugging Face, the GitHub of AI, hosted code that backdoored user devices

• On the Societal Impact of Open Foundation Models

• Will AI sharpen or dull our minds?

• Perceptions in Pixels: Analyzing Perceived Gender and Skin Tone in Real-world Image Search Results

• AI and Surveillance Capitalism

• Avoiding Past Mistakes in Unethical Human Subjects Research: Moving From Artificial Intelligence Principles to Practice

• AI Could Actually Help Rebuild The Middle Class

• Three new books on the social contexts and consequences of algorithms

• AI ethics discourse: a call to embrace complexity, interdisciplinarity, and epistemic humility

Compiled by Leif Hancox-Li