Reuters used Gemini Pro to collect and analyze prison temperature data

"Reuters found that nearly 50% of state prisons across 29 states have partial or no air conditioning in housing units. The Texas Department of Criminal Justice reported that 51 of its 58 facilities fall into this category—the highest number among all states that responded to Reuters."

Locking someone in an unconditioned cell, in Texas, in August heat, is not humane. We don't have to be Scandinavia, but we should expect humane conditions.

> "Reuters found that nearly 50% of state prisons across 29 states have partial or no air conditioning in housing units."

Immediate red flag in my mind reading this line. The only reason I can imagine to phrase it this way is to artificially make the numbers look "bigger" to the reader without outright lying.

That summary is disingenuous at best. Partial/NoAC are 2 distinct categories and Partial doesn't even have a set definition.

There's a pretty big difference between "A single housing units AC at the prison is broken", "AC isn't installed on half our units", and "No AC at all".

> "The Texas Department of Criminal Justice reported that 51 of its 58 facilities fall into this category—the highest number among all states that responded to Reuters"

I'm having trouble reconciling this statement with the data they published to GitHub. I viewed the document on my phone so maybe I missed something while reading it but I can't figure out a way for the numbers on that sheet to match what they wrote here. I thought maybe the data was just updated since the article ran but the source is a PDF document dated 2023 so I doubt that's what's happening here.

https://github.com/reuters-graphics/prison-heat-records/blob...