Shri Shivraj Singh Chouhan Launches Enhanced NMMS App Version with Advanced Security Features for VB-G RAM G

1. At a Glance

• NMMS (National Mobile Monitoring System) is the app used to record workers' attendance under VB-G RAM G, the rural employment scheme that replaces MGNREGA [1][2].
• On 23 Oct 2026, Union Minister Shivraj Singh Chouhan launched a security-focused upgrade of the app. It adds device-level security, AI-based verification, liveness detection, Google Play Integrity checks and safeguards against cloning and tampering [1].
• It was built with technical support from IIT Madras and other technical agencies [1].
• Why it matters for UPSC: it touches several themes at once:
• rural employment guarantees
• digital governance and leakage control
• digital exclusion: workers losing wages when the technology fails
• accountability tools such as social audit, DBT and the e-MB

2. Why in the News

• 23 Oct 2026: the enhanced NMMS version was launched at a press conference in New Delhi [1].
• Last week of Sep 2026: an earlier security update was released. The new version builds on it [1].
• Mar 2026: face authentication was added to NMMS after complaints of fake attendance and of wages being claimed in other people's names [1].
• 1 Jul 2026: VB-G RAM G came into force across all rural areas [1][3].

3. Background & Evolution

• MGNREGA, 2005 guaranteed 100 days of wage employment. The VB-G RAM G Bill was introduced to replace it [2].
• 1 Jan 2023: States and UTs were required to record attendance on NMMS for all works except Individual Beneficiary Works. The rule required two time-stamped, geo-tagged photographs of workers each day [4][5].
• Offline capture was allowed where the network was poor, with the data uploaded once the device reached coverage [4][5].
• 2023 relaxation: the second photograph could be taken any time 4 hours after the first, instead of at a fixed time [4][5].
• 16 Dec 2025: the Viksit Bharat – Guarantee for Rozgar and Ajeevika Mission (Gramin) [VB-G RAM G] Bill, 2025 was introduced in the Lok Sabha [2].
• 1 Jul 2026: the Act came into force across all rural areas, as notified by the Central Government [3][1].
• Mar 2026: face authentication was added to NMMS [1].
• Late Sep 2026: a security update was released [1].
• 23 Oct 2026: the enhanced version was launched, with Play Integrity, liveness detection, region-adapted AI face verification, offline mode and an exception mechanism [1].
• The NMMS launch year (2021) appears only in secondary sources outside the whitelist. Verify it before relying on it.

4. Core Static Facts

Item	Fact
Scheme	VB-G RAM G: Viksit Bharat – Guarantee for Rozgar and Ajeevika Mission (Gramin) [2]
Replaces	MGNREGA, 2005 [2]
Guaranteed days	125 days (up from 100) [2]
Unemployment allowance	Still payable if work is not given within 15 days [2]
In force	1 Jul 2026, all rural areas [1][3]
Nodal ministry	Ministry of Rural Development. The minister holds both Agriculture & Farmers' Welfare and Rural Development; the RD Secretary is Rohit Kansal [1]
Tech partner for the NMMS upgrade	IIT Madras plus technical agencies [1]
Central allocation, FY 2026-27	₹95,692 crore [1]
Total provision including states' share	> ₹1.51 lakh crore [1]
First instalment ("mother sanction")	₹26,126 crore to States and UTs [1]
Worksites	31.23 lakh across > 2.41 lakh Gram Panchayats [1]
Workers / households	> 3.65 crore workers from 2.89 crore households [1]
Person-days generated	~46 crore, of which ~40 crore were recorded by face authentication [1]
States at 100% face authentication	AP, Karnataka, Kerala, Odisha, Tamil Nadu, Telangana, Chhattisgarh, Rajasthan, Uttarakhand (9) [1]

Key terms [1]:

• Liveness detection: checks that the camera sees a real, live worker, not a dummy image or an old photograph.
• AI photo de-duplication: flags suspicious or repeated photos. When it does, the wage list and FTO (Fund Transfer Order) are withheld automatically until checks are done.
• e-MB (electronic Measurement Book): records the work actually done, as physically measured by technical engineers. Wages are calculated from this measurement.
• Exception mechanism: the mate or field worker files a request when face authentication fails, for example for elderly workers or persons with disabilities, or when the device has a problem.

5. Multi-Dimensional Analysis

Economic / Fiscal

• The Centre's ₹95,692 crore is about 63% of the ₹1.51 lakh crore total, which implies a large state share (derived from [1]).
• The first instalment is about 27% of the central allocation (derived from [1]).
• About 46 crore person-days for 2.89 crore households works out to roughly 16 days per household in the first ~4 months (derived from [1]).
• Leakage control matters because public money must be "used transparently and for their intended purpose" [1].

Scientific / Technological

• The upgrade stacks several defences: Play Integrity checks the device and app, then come secure data transmission, anti-cloning and anti-tampering safeguards, liveness detection, and AI face matching tuned to regional differences [1].
• After capture, AI photo de-duplication acts as a further control on wage payments [1].
• The Minister framed cybersecurity as "continuous improvement" against evolving attempts at technological breaches [1].

Social / Equity

• Digital exclusion risk: poor connectivity, failed authentication for elderly and disabled workers, and faulty devices.
• These risks are mitigated by the offline mode, where attendance is uploaded after the muster roll closes, and by the exception mechanism [1].
• The stated principle is that no genuine worker should lose employment or wages because of network or technical problems [1].

Ethical / Governance

• Attendance in the app does not by itself trigger payment. Wages are paid only after layered checks: de-duplication, e-MB measurement, then DBT [1].
• Ground-level accountability comes from random spot inspection teams in districts, social audits and public disclosure [1].
• States must monitor suspicious or unauthorised apps, collect digital evidence promptly, and start criminal proceedings under applicable laws [1].

Administrative / Federal

• Uptake varies across states: only 9 states record 100% of attendance by face authentication [1].
• The Centre sets the technology rules and states enforce them, including prosecutions [1].
• Earlier NMMS rules, such as the twice-daily photographs and the 4-hour gap, show a pattern of revising procedures based on how they work on the ground [4][5].

6. Recent Developments (last 12-18 months)

• 16 Dec 2025: VB-G RAM G Bill introduced in the Lok Sabha [2].
• Mar 2026: face authentication added to NMMS [1].
• 11 May 2026: PIB published an FAQ notifying that the Act takes effect on 1 Jul 2026 [3].
• 1 Jul 2026: VB-G RAM G took effect in all rural areas [1][3].
• Late Sep 2026: NMMS security update released [1].
• 23 Oct 2026: enhanced, security-focused NMMS launched, developed with IIT Madras [1].

7. Prelims Hooks

• VB-G RAM G expands to Viksit Bharat – Guarantee for Rozgar and Ajeevika Mission (Gramin) [2].
• VB-G RAM G guarantees 125 days of work, against 100 under MGNREGA [2].
• The 15-day unemployment allowance trigger is kept from MGNREGA [2].
• The Act took effect in all rural areas on 1 July 2026 [3].
• NMMS stands for National Mobile Monitoring System [1].
• The upgraded NMMS was developed with technical support from IIT Madras [1].
• The upgrade integrates Google Play Integrity [1].
• Liveness detection guards against dummy images and old photographs [1].
• Face authentication was added to NMMS in March 2026 [1].
• Nine states have 100% face-authenticated attendance, including Kerala, Odisha and Uttarakhand [1].
• Central allocation for FY 2026-27 is ₹95,692 crore; the total with states' share is above ₹1.51 lakh crore [1].
• Wages are based on the e-MB, meaning measurement by technical engineers, not on attendance alone [1].
• When AI de-duplication flags an entry, the wage list and FTO are automatically withheld [1].
• NMMS twice-daily geo-tagged attendance has been mandatory since 1 Jan 2023, with Individual Beneficiary Works exempt [4][5].

8. Who Pays When the App Wrongly Flags a Genuine Worker

• A wrong flag stops the wage first and checks later
• When AI de-duplication marks a photo as suspicious, the wage list and FTO (Fund Transfer Order, the instruction that sends money to the worker's bank) are held back automatically [1].
• A real worker whose photo is flagged by mistake waits for wages until an official clears the case. The worker did nothing wrong.
• Being paid late already hurts this group. In 2022 a parliamentary Standing Committee found that delay compensation (extra money owed to workers when wages come late) was not paid in most places in the country [11].

• So the cost of a machine error falls on the poorest person in the chain, not on the system that made the error.

• Small error rates still mean very large numbers

• About 40 crore person-days were recorded through face authentication [1].
• Illustration only: if just 1 in 100 of these were wrongly flagged, that is about 40 lakh days of work whose wages are stuck.

• The release gives no figure for how many flags turned out to be wrong [1]. Without that figure, nobody can tell whether the system is catching fraud or mostly delaying honest workers.

• Earlier app rules have already cost workers their wages

• In February 2024 the Standing Committee on Rural Development reported that some workers could not mark attendance on NMMS, and so were not paid for work they had done [6].
• It also noted that many workers struggle with internet connectivity and do not have smartphones [6].
• In December 2022 a Rajya Sabha question asked whether the government would give smartphones to workers who were being forced to buy them for NMMS attendance [8].
• Each new layer of security adds one more place where a genuine worker can fail the check.

9. Why the Safety Net for Failed Face Checks Is Still Thin

• Face checks do not fail equally for everyone
• The World Bank's ID4D (Identification for Development) programme notes that some groups of people have higher failure-to-acquire rates, meaning the machine cannot capture a usable biometric from them [13].
• The release itself names elderly workers and persons with disabilities as the people who need the exception route [1].

• So the people most likely to be rejected are often the people the scheme most needs to protect.

• The exception route depends on the field worker, not the worker's own right

• Under the new system, the mate (the local person who supervises the worksite and marks attendance) files the exception request [1].
• The release does not say how fast a request must be decided, who decides it, or how a worker can appeal a refusal [1].
• The World Bank's ID4D guide says exception handling must be open and practical, and must come with a grievance redress system (a formal way to complain and get a decision) [12].

• Without time limits and an appeal route, the exception becomes a favour from the mate, not an entitlement of the worker.

• Payments already have a backup route — attendance should too

• For wage payments, ABPS (Aadhaar Based Payment System, which routes money using the Aadhaar number) has been mandatory since 1 January 2024 [7].
• If an ABPS payment fails, the money can still go through NACH (the normal bank-account-based route) [7].
• Face-based attendance has no equally clear, routine backup. It only has a case-by-case exception [1].

10. The Strongest Case for Face Checks, and What It Misses

• The case for the app is real
• Face authentication was added in March 2026 because of complaints of fake attendance and of wages taken in other people's names [1].
• The government says ABPS helps make sure only people with verified identities get wages [7].
• Even Parliament's committee, which has criticised the app, recommended photo and video checks of works to stop false claims [10].

• So the question is not whether to use technology. It is how much risk to put on the worker when it fails.

• Concede this: ghost workers do take real money

• A fake name on the muster roll (the attendance register) takes a share of the wage bill that belongs to genuine workers.

• Face matching and liveness detection directly attack this kind of fraud [1].

• But the bigger leak may not be in attendance at all

• Wages are worked out from the e-MB, the measurement of the work actually done [1].
• If an engineer overstates how much earth was dug or how much wall was built, the extra money leaks out even if every face is genuine.

• The app checks who came. It does not check how much work was really done. That is why the committee asked for photo and video checks of the works themselves [10], and why social audits still matter.

• The same committee warned against forcing a tool before it is ready

• In 2024 it found 14% of job card holders were not Aadhaar-enabled (as of November 2023) [6].
• It recommended that Aadhaar should not be made mandatory until the system was made foolproof [6].
• The same logic applies to face authentication: a check that blocks payment should be proven accurate first, not fixed after workers lose wages.

11. What the Ministry Should Do Next

• Ministry of Rural Development: publish how often the app gets it wrong
• Release state-wise figures every month: how many photos were flagged, how many were later cleared as genuine, and how long clearing took.

• The release gives only success numbers, such as 40 crore face-authenticated person-days [1]. Failure numbers are needed to judge if the system is fair.

• Ministry: set a deadline for clearing flagged wages, and pay compensation automatically

• A flagged FTO should be cleared or rejected within a fixed number of days.

• If the delay comes from a wrong flag, delay compensation should be paid without the worker having to ask. The 2022 Standing Committee found this compensation was mostly not paid [11].

• Ministry: give attendance a routine backup, like payments already have

• ABPS has NACH as an ordinary fallback when it fails [7].
• Attendance should have a similar standard route (for example, verification by the Gram Panchayat or block officer), not only a one-off exception filed by the mate [1].

• This follows the World Bank ID4D guidance that people must have another way to prove identity when biometrics fail [12].

• States: let workers appeal, not only mates

• A worker who is refused an exception should be able to complain directly, with a written decision, through a grievance redress system [12].

• Ministry and states: point the new money at measuring work, not only faces

• Spend part of the security effort on checking the e-MB, through photo and video checks of the works themselves, as the Standing Committee recommended [10].
• Back this with stronger social audits, the check the Minister himself cited [1].

12. Anchors for Answers

• Data: ~40 crore of ~46 crore person-days recorded through face authentication in FY 2026-27 so far [1]
• Data: 14% of MGNREGA job card holders not Aadhaar-enabled as of November 2023 [6]
• Data: NMMS real-time, geo-tagged attendance started on 21 May 2021 [9]
• Report/Committee: Standing Committee on Rural Development and Panchayati Raj, February 2024: workers unable to mark NMMS attendance went unpaid; Aadhaar should not be mandatory until foolproof [6]
• Report/Committee: Standing Committee, February 2022: delay compensation not paid in most places; recommended 150 days of work [11]
• Report/Committee: Standing Committee 2024-25: photo and video verification of works to stop false claims [10]
• Law/Case: K.S. Puttaswamy v. Union of India (privacy as a fundamental right; Aadhaar judgment); Digital Personal Data Protection Act, 2023 (face data of workers)
• Comparison: World Bank ID4D guidance: when biometrics fail, people must have another way to prove identity, with transparent exceptions and grievance redress [12]
• Scheme: ABPS in MGNREGA, mandatory since 1 January 2024, with NACH as the backup payment route — a model for a backup attendance route [7]

13. Mains Relevance

• GS-II:
• Government policies and interventions
• Welfare schemes for vulnerable sections
• E-governance, transparency and accountability

• Social audit

• GS-III:

• Inclusive growth and employment

• Cybersecurity and the use of AI in governance

• GS-IV: probity in governance; the tension between preventing fraud and protecting entitlements.

• Question stems: 1. "Technology-driven verification in welfare delivery can curb leakages but risks excluding genuine beneficiaries." Examine this with reference to face authentication under VB-G RAM G. 2. Discuss how layered controls (AI de-duplication, e-MB, DBT and social audit) strengthen accountability in rural employment programmes. What are their limitations? 3. Critically compare VB-G RAM G with MGNREGA in terms of entitlements, fiscal federalism and monitoring architecture.

14. Related Topics to Study Next

• MGNREGA, 2005: the rights-based framework that VB-G RAM G replaces.
• Social audit (Gram Sabha, MGNREGA Section 17 framework): the accountability layer the Minister cited.
• DBT and the JAM trinity: the payment rail behind traceability.
• Aadhaar-based authentication and the Puttaswamy (2017/2018) judgments: the privacy and exclusion questions raised by face authentication.
• Digital Personal Data Protection Act, 2023: governs the biometric and face data NMMS collects.
• AI in governance (IndiaAI Mission): context for AI-based verification.
• Farmer ID / AgriStack: the related digital registry drive mentioned in the same release [1].
• Fiscal federalism in centrally sponsored schemes: the Centre-state cost split.

15. NCERT Concepts to Revise

• Mahatma Gandhi National Rural Employment Guarantee Act 2005 (Polity): The 100-day rights-based law that VB-G RAM G replaces; the NMMS app monitors its successor.
• Social audit (Polity): Ground-level check by beneficiaries that catches fraud the app cannot, such as inflated work measurements.
• Aadhaar (Polity): Biometric authentication and its exclusion problems, such as the unwaged workers who were not Aadhaar-enabled, parallel face authentication here.
• JAM trinity (Polity): Jan Dhan, Aadhaar and mobile form the DBT payment rail that wages travel on after verification.
• Digital Personal Data Protection Act 2023 (Polity): Governs how workers' face and biometric data collected by the app must be handled.
• E-governance (Economy): NMMS is technology-driven service delivery; it shows both the leakage-control gains and the exclusion risks.

16. Common Errors / Trap Areas

• Ministry: the scheme is run by the Ministry of Rural Development, not the Agriculture Ministry, even though Chouhan holds both portfolios [1].
• Attendance ≠ payment: wages follow e-MB measurement, not app attendance alone [1].
• Dates: face authentication came in Mar 2026, the scheme took effect on 1 Jul 2026, the security update came in Sep 2026 and the enhanced app in Oct 2026. NMMS photo attendance has been mandatory since 1 Jan 2023. Don't mix these up [1][3][4].
• Days: VB-G RAM G guarantees 125, not 100 [2].
• Offline mode: offline attendance is uploaded after the muster roll is closed. It is an allowed route under the prescribed procedure, not a loophole [1].