Next generation XBRL: first public working draft now available!
The XBRL Standards Board has approved publication of the first public working draft (PWD) of Project Tavi, the next generation XBRL specification. It sets out a simpler approach to modelling and digitising reports of all kinds, and we would like your help to shape it.
“Project Tavi” is a temporary name; a more formal one will be announced at Data Amplified 2026 in a few weeks. As an initial PWD, the specification remains entirely subject to change, and some parts have yet to be written. We are publishing it because it is sufficiently developed to be useful, and because broader input now will help us get the direction right.
The central change is a simplified data model, represented for now in JSON, with far fewer moving parts. Ordinary developer tools can work with it, and a capable general purpose AI tool can read taxonomies directly, answer questions about it and use it to work with the associated data.
The fundamentals do not change. Concepts, dimensions and data constraints remain, because they carry the meaning of reporting. The draft also streamlines terminology and folds the registries for units, data types and link roles into taxonomies themselves, and it comes with example models and draft conformance suites to experiment with.
Importantly, the XBRL 2.1 stack is not being deprecated, now or on any timetable. Existing taxonomies and filings remain valid, there is no migration deadline and no one will be required to adopt the new framework {Ed: XBRL International couldn’t, and wouldn’t force that in any event!}. For now the value lies in understanding the direction of travel and helping to shape it.
We welcome feedback on whether this is heading the right way, what is missing and what could be trimmed.
Read more here, and send feedback to oim-feedback@xbrl.org.
