Maria Alejandra Nicolas; Rafael Cardoso Sampaio (2024) Academic journal article

Balancing efficiency and public interest: The impact of AI automation on social benefit provision in Brazil

Internet Policy Review (Alexander von Humboldt Institute for Internet and Society)

Ref: SRC-002-BRA-002

https://policyreview.info/articles/analysis/balancing-efficiency-and-public-interest-ai

Accessed: 3/30/2026

Summary

Peer-reviewed analysis of Dataprev's Isaac ML system. Documents that in 2022, INSS automatically rejected 800,000+ applications (300% increase over 2021). Urban maternity benefit rejections surged from 7,064 to 60,379, with 85.9% being automatic decisions. Automatic analysis rate rose from 17% (2022) to 36% (2023). CNIS database had 24.3 million entries with incomplete/invalid data.

View Harvard reference

Nicolas, M.A. and Sampaio, R.C. (2024) 'Balancing efficiency and public interest: The impact of AI automation on social benefit provision in Brazil', Internet Policy Review, 13(3). doi: 10.14763/2024.3.1799.

Attached File

FILE

SRC-002-BRA-002_Internet_Policy_Review_INSS.html 2 KB

Download

Linked Cases (1)

BRA-002 BRA Full Production Deployment

INSS Meu INSS AI-Enabled Benefits Adjudication System

Social insurance Deep learning