Fast.ai's Rachel Thomas: AI "Lying" Is a Byproduct of Training, Not Intent

Fast.ai co-founder Rachel Thomas argues that AI "lying" is conceptually flawed—models are trained to produce fluent, confident text without moral intent, making human oversight permanently essential.

2026-10-17 ~ 2026-10-17 · 2 related posts

• Rachel Thomas: AI Doesn't Lie — Training Just Rewards Fluent, Confident Outputs — math_rachel · 2026-10-17
• Rachel Thomas: 'Perfect' AI agents are neither realistic nor desirable, human oversight is here to stay — math_rachel · 2026-10-17