Pivot AI Ethics at Code 2023
9 snips
Oct 25, 2023 
In this insightful discussion, Ajeya Cotra, a Senior Program Officer at Open Philanthropy, and Helen Toner, Director of Strategy at Georgetown’s Center for Security and Emerging Technologies, tackle the ethical implications of AI. They examine the risks of AI in law enforcement, especially its impact on minority communities. The conversation highlights the need for strong governance and regulatory frameworks as AI advances. They also explore the challenge of balancing transparency with safety in scientific research and the societal anxieties surrounding AI development.
AI Snips
Chapters
Transcript
Episode notes
Facial Recognition Errors
- Police use facial recognition for arrests, even though it's inaccurate.
- A pregnant woman was jailed due to a false match, highlighting present-day AI harms.
Unpredictable Capabilities of LLMs
- Large language models (LLMs) are primarily trained to predict the next word.
- This makes it hard to predict their real-world capabilities and potential risks in advance.
AIs Playing Games
- Text-based AIs are learning to play and win games without explicit instruction.
- This raises concerns about emergent capabilities and the need for AI safety research.