ChatEDU – The AI & Education Podcast

Beyond the Vision: Building a Policy Infrastructure to Meet the AI Moment | Ep. 88

10 snips
Dec 12, 2025
The hosts dive into the new voice mode of ChatGPT, revealing its capabilities and risks. They analyze a compelling MIT study that reveals AI's impact on jobs, debunking clickbait headlines. Comparing ChatGPT and Gemini, they explore personalized job forecasting. The safety of AI chatbots in schools is examined, advocating for a new protocol to address student disclosures. Finally, they propose four key policy shifts to adapt education for the AI age, emphasizing the need for durable skills and coordinated efforts across stakeholders.
Ask episode
AI Snips
Chapters
Books
Transcript
Episode notes
ANECDOTE

Voice Mode Hallucinated Golf Club Images

  • Matt demos ChatGPT's advanced voice mode and it produced inaccurate images for Torrington Country Club.
  • Jonathan confirms the visuals were plausible but wrong based on his personal knowledge.
INSIGHT

AI Exposure Is Nuanced Not Apocalyptic

  • The MIT Iceberg Index shows many jobs contain skills overlapping with AI, but headlines exaggerated mass replacement.
  • Jonathan and Matt stress complex, human-facing roles have much smaller AI overlap and remain safer.
INSIGHT

Errors Multiply Across Multi-Step Work

  • Multi-step tasks multiply error rates so agentic systems often fail on complex workflows.
  • Jonathan warns small per-step errors compound into large end-to-end failures.
Get the Snipd Podcast app to discover more snips from this episode
Get the app