For AI & data teams
Consented, rewarded and traceable human data, including the languages the open web leaves out.
Provenance is the first question in any diligence or licensing talk.
Preference data and model evaluation need real people with real context, not synthetic judgements.
Most languages and populations are thin or absent in web-scale data.
Contributors join under published terms, and each dataset records where it came from.
Every answer is hashed the moment it arrives and anchored on the blockchain. Your counterparty can verify it without an account.
Field one instrument in the languages your contributors think in.
One real person per account, with no ID document. Exports never carry names or contacts.
A reward in SURDA for each approved answer, even where the usual payment services do not reach.
Reach your studies through the API, and download datasets you buy as CSV.
Verify any dataset on the blockchain yourself. No account needed.
Verify on the blockchain →The blockchain record shows a dataset is unchanged and its contributors were paid. Whether a sample represents a population depends on the study design.
Who you need to hear from, how many, and where.
