557: LLMs: What Data Viz Teaches Us

Mike & Tommy dive into what data visualization can teach us about critically evaluating LLMs, exploring why fluent, polished AI output can be just as misleading as a beautiful but inaccurate chart. They break down the tension between AI speed and data quality, tackling how BI teams can build a practical framework around accuracy, security, and sovereignty to use these tools without lowering their standards.

Read the article that inspired this episode: https://nightingaledvs.com/critically-evaluating-llms/

News links from this episode:

https://www.meetup.com/chicagolandpowerbi/events/315787998/?eventOrigin=group_upcoming_events

https://community.fabric.microsoft.com/blog/fbc_pbiupdatesblog/power-bi-august-2026-feature-summary/5348434

https://community.fabric.microsoft.com/t5/Fabric-Updates-Blog/Bringing-agentic-warehouse-development-to-Fabric-Data-Warehouse/ba-p/5360190

https://community.fabric.microsoft.com/blog/fbc_pbiupdatesblog/the-ai-semantic-layer-you-probably-already-have/5360197

Get in touch:

Send in your questions or topics you want us to discuss by tweeting to @PowerBITips with the hashtag #empMailbag or submit on the PowerBI.tips Podcast Page.

Visit PowerBI.tips: https://powerbi.tips/

Watch the episodes live every Tuesday and Thursday morning at 730am CST on YouTube: https://www.youtube.com/powerbitips

Subscribe on Spotify: https://open.spotify.com/show/230fp78XmHHRXTiYICRLVv

Subscribe on Apple: https://podcasts.apple.com/us/podcast/explicit-measures-podcast/id1568944083‎

Check Out Community Jam: https://jam.powerbi.tips

Follow Mike: https://www.linkedin.com/in/michaelcarlo/

Follow Tommy: https://www.linkedin.com/in/tommypuglia/