BrazilPublic sector2020 – ongoingControladoria-Geral da União (CGU)
Faro AI triage of citizen complaints
CGU reported that Faro analyzed 4,024 complaints and automatically handled 610 in 2022
Problem
CGU needed to assess whether citizen complaints contained enough information for investigation. Incomplete submissions required further information or closure.
Approach
CGU's Strategic Information Directorate developed Faro using natural-language processing and machine learning. The system scored complaints submitted through Fala.BR. In 2022, it automatically archived anonymous submissions lacking minimum information and requested additions to other incomplete complaints.
Results
Takeaway
The reported automation concerned complaint eligibility and follow-up. It did not measure the accuracy of subsequent investigations.
Categories
- Size
- Large enterprise
- Industry
- Public sector
- Business function
- Production and quality
- Innovation type
- Process innovation
- AI technology
- Predictive analytics and ML
- Adoption
- Built in-house
- Evidence
- Company's own claim
Sources
Source summaries
CGU described Faro's NLP and machine-learning triage. It reported 4,024 complaints analyzed and 610 handled automatically, comprising 464 archived submissions and 146 requests for further information.
CGU explained Faro's complaint scoring and its use in preliminary assessment. Its comparison of 2020 and 2021 activity established that the system was already operating in 2020.
Full details are in the original sources.
