Face recognition log-ins cut health staff absences in Andhra Pradesh

The World Bank's evaluation office reports that a facial recognition system for logging health staff in and out of shifts reduced absenteeism in Andhra Pradesh and cut salary spending on absent staff.

  • 2 for an organization

    The evaluators report reduced absenteeism and lower salary spending on absent staff across the state's health workforce, and the Bank's report puts absenteeism at a negligible level, but neither gives a before and after figure.

Depth runs from 1, a convenience or time saved, to 5, a life or a livelihood preserved.

Status
Reported benefit.
Evidence

foundstudiedreportedclaimed

An independent descriptive report supports the benefit.

Size

No magnitude yet: duration and distribution are not documented.

The system’s part
Not isolated. The evaluators credit the facial recognition system with reduced absenteeism and salary savings, relaying the Bank team's efficiency analysis. Neither document gives absenteeism figures, separates the system from the zero-vacancy policy and other staffing measures or describes the recognition model. Neither shows that the system learns from data.
Not yet checked
LimitsCosts and harmsLearns from dataLater outcome

Download data Cite Suggest a correction

Benefit type
Improved Organizational Efficiency
System
Facial recognition attendance system with geo-location for health staff in Andhra Pradesh's public health facilities (not named in the review)
Made by
Not named in the review
Used by
Andhra Pradesh Department of Health and Family Welfare (DoHFW)
Where
Andhra Pradesh, India
Stage
In routine use
Date
Kind
Aggregate result
Broad benefit
Operational improvement (enhanced productivity)
Collections
Scrutiny and findings
Last revised
1 November 2026

Sources

  1. Regulatory document, by authors with no tie to the system. Context or contrary evidence.

    MediaNama (relaying The Times of India's report of a National Medical Commission finding) (2024). Andhra Pradesh medical colleges to switch from facial recognition to biometric attendance, Article body, paragraphs 1 and 4.

    Under the facial recognition attendance system, the NMC found some doctors would leave the college premises after marking their attendance through their mobile to attend to their private practice.

  2. Report, by an evaluator that the makers or users commissioned.

    World Bank. Implementation Completion and Results Report: Andhra Pradesh Health Systems Strengthening Project (P167581). ICR00457, 30 October 2025. The Bank's own completion report, which the Independent Evaluation Group reviewed.

    The new facial recognition system for health personnel to log in/out at the start and end of each shift that includes geo-location and shows where each person is at all times during their shift, has reduced absenteeism to a negligible level.

  3. Descriptive report, by authors with no tie to the system.

    World Bank Independent Evaluation Group. Implementation Completion Report (ICR) Review: Andhra Pradesh Health Systems Strengthening Project (P167581), ICRR0024843, 5 May 2026. On this point the review restates the Bank's completion report.

    a new facial recognition system for health staff to log in and out of each work shift, which reduced absenteeism and led to cost reductions associated with paying salaries to absentee staff

Details

AI system

Workflow

Health staff logged in and out of each shift through a facial recognition system with geo-location. The state pursued a zero-vacancy policy supported by digitized attendance systems.

Follow-up

Conditions for replication

Facilities must have devices and connectivity for the recognition system at every shift. Managers must act on the attendance records, for example by withholding pay for absences.

Entry details

Sector

Healthcare (workforce management).

Purpose

An unnamed supplier built the facial recognition system for the Andhra Pradesh health department so that staff log in and out of each shift, to reduce absenteeism and the salary cost of absent staff.

31 more questions not yet checked

model, release, provider, role, configuration, inputs and outputs, autonomy, human oversight, AI contribution, beneficiaries, reach, distribution and access, duration, every measurement question, every money question, ownership and control, dependencies, cross-border connections, geopolitical implications, replication attempts, availability, institutional decision, grounds, action and result, success criterion, every evidence question, observed or intended, alternative assessments, related entries, other identifiers, category mappings.

Cite this benefit

Peter H. Vartanian. (2026). Face recognition log-ins cut health staff absences in Andhra Pradesh (AIBD-1174). Artificial Intelligence Benefits Database, edition of 2 November 2026. https://petervartanian.xyz/aibd/records/AIBD-1174.html (archived at https://doi.org/10.7910/DVN/AADSII, version 9.0)

BibTeX Data Field definitions