Awkward Silences #187 - Throwback: Harnessing AI For Better Insights with George Whitfield of MIT and Google
33 snips
Apr 14, 2026 George Whitfield, MIT lecturer, Google software engineer, and founder of Find Our View, applies AI to make sense of large-scale qualitative data. He explores where language models help versus when human judgment is essential. The conversation touches on preserving context in transcripts, human-in-the-loop workflows, hypothesis-driven prompts, and a multimodal future for richer insights.
AI Snips
Chapters
Books
Transcript
Episode notes
Two Distinct Challenges Of Scaling Qualitative Research
- Large qualitative sets create two challenges: immediate digestion of each interview and seeing emergent patterns across dozens of conversations.
- George Whitfield describes the short-cycle need to capture per-interview insights and the long-cycle task of finding cross-interview trends when you hit 10–100 interviews.
Selling Insights To The New York Times Sparked The Pivot
- George started in consumer insights and sold a product to the New York Times before pivoting to user research tools.
- He noticed large conversation volumes lost personal context, inspiring FindOurView to preserve voice at scale.
Prime AI With Your Hypotheses Before Analysis
- Inject your research intentions and hypotheses into the AI prompt so outputs validate or invalidate what you set out to test.
- George explains FindOurView runs hypotheses through transcripts and bookmarks direct quotes as evidence to reduce hallucination.
