Pietro Miotti's 48 AI agents 'find' a Maya site already in a journal

The episode began as a personal side project: Miotti, traveling in Peru, wondered what historic explorers might do with modern tools, settled on lidar, and set 48 AI agents to comb NASA lidar data after finding Peru poorly covered by public lidar. The agents flagged numerous locations, including a 'major village' that did not appear in existing records, and Miotti reached out to INAH to check whether it was already known. When INAH said it had no record, Miotti treated that as confirmation and publicized the finding, stressing he is not an archaeologist and that fieldwork would be needed to verify it. The problem, as archaeologists pointed out, is that a government database's silence is not proof of an unknown site: the site had been documented in an academic journal more than a year earlier, and Ivan Šprajc, a coauthor of that paper, noted that INAH's lack of official knowledge does not mean archaeologists have not spotted or explored a site. Sarah Parcak, an archaeology professor, stressed that this is why peer review exists, and other archaeologists criticized the project for spreading misinformation and taking credit from researchers who have used digital tools for decades. Miotti has since updated his post to credit the archaeologists and said he did not expect the attention, while Elizabeth Graham of University College London argued the bigger obstacle is funding for excavation and mapping, not the number of sites 'discovered.'