Update (11 November): it turns out the site had already been mapped in Estrada-Belli et al. 2025. The agents missed it while going through the literature, so they rediscovered it rather than discovered it (it just wasn't in INAH's register yet). Still work to do then!

Not undiscovered. Bad AI output. This is, ultimately, an article about slop.

Interesting article, but only surprising part is that this hasnt been done before (mass analysing the Lidar Data). This seems like something that could have been done with much less compute and more specialized machen Learning algorithms already years ago. I mean most shapes that point to old buildings under the foilage are not exactly suddle in the Lidar Data.

I hate the phrase "my AI agents"