# Tilting the Wildfire Odds

Page: https://stenobird.com/podcast/the-nfpa-podcast-1063918/tilting-the-wildfire-odds
Text version: https://stenobird.com/podcast/the-nfpa-podcast-1063918/tilting-the-wildfire-odds.md
Podcast: [The NFPA Podcast](https://stenobird.com/podcast/the-nfpa-podcast-1063918)
Published: 2025-11-11T15:33:00+00:00
Episode link: https://nfpa.libsyn.com/tilting-the-wildfire-odds
Audio file: https://traffic.libsyn.com/secure/nfpa/EP126_Gollner_AI_FINAL_-17_LUFS.mp3?dest-id=12612
Processing state: not_requested
JSON: https://stenobird.com/v1/public/podcasts/the-nfpa-podcast-1063918/episodes/tilting-the-wildfire-odds
Duration seconds: 3142

## Resource

In even the most destructive wildfires, some structures survive untouched while everything around them is incinerated. It begs the question: What factors are most responsible for determining if a home is destroyed or survives during these events? A team of researchers spent years trying to answer that important question. They gathered massive troves of data from California's most destructive wildfires, then used artificial intelligence machine learning to analyze what factors make the biggest difference in home survivability. On today's podcast, we talk to wildfire researcher and study co-author Michael Gollner about what the team found out, how they did it, and how this valuable information will be used. We also discuss whether AI is changing the way we study and think about wildfire. LINKS: Read the wildfire research paper we discussed in this episode. Read the NFPA Journal article about this project Learn more about the Firewise USA program at NFPA

## Actions

- request_transcript: `POST https://stenobird.com/v1/public/podcasts/the-nfpa-podcast-1063918/episodes/tilting-the-wildfire-odds/transcription-requests` — Idempotently request low-priority transcript generation for this episode.
- read_markdown: `GET https://stenobird.com/podcast/the-nfpa-podcast-1063918/tilting-the-wildfire-odds.md` — Read the agent-friendly Markdown representation of this episode resource.

A page view does not enqueue transcription. Agents should invoke `request_transcript` explicitly when they need this episode processed.

## Transcript

Full transcripts are not published on public pages unless there is a clear rights basis.