Education Research Reading Room ERRR #102. Benjamin Riley on the dangers of AI in Education
40 snips
May 7, 2025 Benjamin Riley, founder of Cognitive Resonance and former head of Deans for Impact, dives deep into the perils of AI in education. He highlights the necessity for caution in applying AI technologies, discussing their unpredictable nature and ethical implications. Riley contrasts the evolution of educational systems in Australia and England, emphasizing the role of cognitive science and collaborative efforts. The conversation also touches on the societal risks of AI, including its potential to undermine traditional teaching and its connection to authoritarianism.
AI Snips
Chapters
Books
Transcript
Episode notes
AI's Surprising Emergence
- AI emerged as a surprise with ChatGPT's release in 2022, revealing unexpected capabilities.
- Benjamin Riley connected AI's simulation of cognition to his interest in human thinking and education.
AI Predicts, Doesn't Reason
- Large language models rely on statistical predictions, not true human reasoning.
- Understanding AI's probabilistic processes helps demystify how it generates responses.
AI's Bias Exposed by Riddle
- A riddle demonstrates how AI's biases cause illogical answers about a surgeon's identity.
- ChatGPT often incorrectly assumes the surgeon is the mother, even when stated she is dead.
