The updated Global Embedded Metadata Graph, including automatic JSON-LD correction and enhanced handling of nonstandard character escapes will be launching later this week, together with our 2018-present historical backfile!

We're tremendously excited at the vast new landscape of research questions this dataset will make possible, including reaching beyond GDELT's 65 live-translated languages to take the first steps towards metadata-based exploration of the full set of 152 languages GDELT monitors. We're especially interested in new opportunities for misinformation research this dataset opens up, including access to fact checking annotations and embedded provenance information that can help contextualize coverage.