The Family History AI Show EP36: Browser Wars Heat Up, What AI Can Learn from AOL, Anthropic's Speedy New AI Model, Simple Path to Better Prompts
Nov 3, 2025
Discover how AI-enhanced browsers are revolutionizing genealogical research and making data collection effortless. Learn lessons from AOL's rise that can guide AI's broader adoption today. The podcast also dives into Anthropic's lightning-fast Haiku 4.5, perfect for simple tasks, and discusses how to effectively use AI for script extraction. Enjoy insights on new AI features in spreadsheets and exciting advancements in handwritten text recognition that promise to transform how we handle archival records.
AI Snips
Chapters
Transcript
Episode notes
Prefer Small Models For Routine Tasks
- Default to smaller, faster models (Haiku) for summarization, extraction, and routine generation tasks.
- Switch to larger reasoning models (Sonnet/Opus or GPT‑5) only for deep analysis or multi‑document comparisons.
Right‑Sized Models Improve Efficiency
- Smaller models are closing the gap for many everyday tasks, making large models reserved for complex reasoning.
- Using the right model size reduces waiting time and cost without sacrificing practical outcomes.
Build Reusable Prompts Using Stepwise Steps
- Use stepwise prompting: identify the image, list extractable facts, then ask a strong model to build a reusable structured prompt.
- Save and test that prompt repeatedly to process similar records in seconds.