AI Transparency Reports Promise More Than They Deliver

AI Ethics with Fexingo: Bias, Safety, and Responsible Artificial Intelligence · May 23, 2026 · 11m · transcript

Lucas and Luna examine the growing trend of AI transparency reports, which major tech companies like OpenAI, Google, and Meta publish to disclose how their models are built, tested, and monitored. But do these reports actually hold companies accountable? Lucas drills into the case of Anthropic's Model Card for Claude, showing that even the most detailed reports omit critical information about training data sources and red-teaming methodologies. Luna pushes back on whether regulation can mandate useful disclosure without revealing trade secrets. The episode closes by asking whether the next…