Trustingsocial

Trusting Social Philippines — Real-time application-moment risk signals derived from computer vision on borrower onboarding: the Vision Score (a 0-100 score) computed from biometric and behavioural signals captured during loan or account application, sold as an independent present-moment risk read that complements historical bureau scores. The underlying asset is an accumulated corpus of onboarding biometric captures (selfie/video frames, liveness and document images) joined to the resulting risk score AND, crucially, to downstream loan performance flowing back from lender clients — a score-t…

Trustingsocial offers (Alternative, Sentiment, Fundamental) — Not published for this entity. Bounding inference: the market it scores into is large and fast-growing (BNPL and personal-loan apps booked ₱853 billion in 2024; consumer loans ₱3.4 trillion in H1 2025), so an active application-scoring vendor plausibly processes millions of applications per year, each leaving an onboarding capture, a Vision Score and — where clients feed performance back — an outcome. Request Philippine application volume, biometric capture count, and how many applications have a linked repayment outcome..

Thin-file and first-time-borrower underwriting (its explicit positioning), application-moment fraud and risk signal fusion with bureau scores, BNPL and personal-loan-app portfolio quality monitoring, PD model development and challenger-model benchmarking on Philippine consumer credit, deepfake/identity-fraud screening at onboarding, model-validation reference sets (published Gini benchmark implies a labelled evaluation set), and inclusion analytics for the segment the bureau's historical models serve poorly.

The data is with 5 years of history.

Coverage spans APAC, US; Non-Bank Credit, Application Software; alternative, sentiment, fundamental; equities.