Race and AI: Algorithmic Bias and Justice

Foundations of Algorithmic Justice

• 💡 The discussion builds on pioneering work by Dr. Latana Sweeney, Dr. Sophia Noble, Dr. Ruha Benjamin, and Dr. Timnit Gebru on algorithmic bias and auditing.
• 🧠 Dr. Joy Buolamwini's work, including the Algorithmic Justice League, continues to address the ethical implications of AI.

Generative AI and Creative Professions

• ⚠️ Generative AI poses challenges to artists, with concerns about job displacement and the creation of art that lacks soul or rigor.
• 🎨 The "Four C's" campaign (Consent, Compensation, Credit, Control/Agency) advocates for ethical practices when AI systems use artists' work.
• 💡 Many AI-generated creations are seen as "regurgitative AI," often drawing from human-created content without proper permission or compensation.

The Perils of Facial Recognition

• 🚨 The accelerated adoption of AI, particularly in military applications and surveillance infrastructure, raises significant ethical and safety concerns.
• 🔍 Facial recognition technology (FRT), like that used by the TSA, often operates without informed consent and can lead to negative treatment for those attempting to opt out.
• ❌ There's a critical distinction between face detection (identifying a face) and facial recognition (linking a face to an identity), with the latter leading to false arrests and misidentification.

Combating AI Harms and Misinformation

• 🧠 Techno-chauvinism describes the tendency to over-credit machines, leading to a surrender of critical thinking to AI systems.
• ⚠️ Deepfake technology presents severe dangers, including the widespread creation of non-consensual explicit imagery and the potential to blur truth and invalidate historical records.
• ✅ The concept of provenance is crucial for tracing the origin of AI-generated content, offering a potential solution for accountability and ethical use.

Advocacy and Future Directions

• ⚖️ State-level legislation, such as Illinois' BIPA, demonstrates that resistance works in preventing invasive AI uses, highlighting the need for federal protections.
• 💡 The idea that AI's negative impacts are "inevitable" is a marketing narrative, and humans retain the power to steer and shepherd these tools ethically.
• 🎬 The documentary "Coded Bias" humanizes AI harms and continues to tour globally, fostering conversations about respect and justice in AI development.