A one-district, 90-day pilot proposal — measured on a single honest number.
The CAG's performance audit of CSC 2.0, tabled in Parliament this April, says what everyone in the network already knows, with numbers attached: 86% of rural VLEs earn under ₹500 a month against a ₹50,000 setup — an eight-year payback. 35–47% of centres were inactive across 2018–23. Nearly half transact fewer than 100 times a year. e-District services are unavailable at 63.64% of centres, PMFBY at 48.48%, and 54.23% of citizens who walked into a CSC got no service at all. Three-quarters of centres display no rate list — the mechanical cause of the overcharging the audit censures.
The audit's instructions to CSC-SPV — revive inactive centres, widen service integration, benchmark the top earners, end unauthorised training charges — are not a critique of the network. They are a mandate to fix a capability gap. That gap is precisely what we build for.
The most telling comparison in the audit is same-rails, same-villages: CSC Telangana's 72 lakh transactions against MeeSeva's 912 lakh; operator incomes of ₹261 a month against ₹3,000–10,000. The difference is not the villages, the rails, or the operators. It is the breadth of what the operator can deliver. A VLE is authorised for 400+ services and typically delivers about 15.
A Kannada-first capability layer for VLEs — an assistant that lives where the VLE already lives (WhatsApp — no app to install, no portal to learn) and widens what each centre can confidently deliver. Three services first, chosen from the audit's own findings:
e-District, done everywhere. The paperwork services missing from 64% of centres — certificates, documents, applications at ₹50–200 each — are a knowledge problem, not a licensing problem. The assistant walks a VLE through any e-District service in Kannada, step by step, including the official rate list shown to the citizen — three audit findings (availability, income, overcharging) addressed by one behaviour.
PMFBY claims, inside the 72-hour window. The single largest cause of rejected crop-insurance claims is missing the non-negotiable 72-hour crop-loss reporting window, followed by wrong bank/crop/Aadhaar details. When a weather event hits a district, the assistant tells each VLE which of their farmers must report, what is needed, and by when. Every season, the VLE becomes the person who got the village paid.
Insurance, properly served. Rural insurance access is where the knowledge economics are strongest for a VLE — and IRDAI's own village-level distribution push is building the demand now. The assistant supports the VLE through needs-assessment and the RAP pathway in Kannada, so the centre earns from knowledge work, not just throughput.
We build accountable AI systems; this network, of all networks, deserves one. Five commitments, by architecture rather than policy:
We are a Bengaluru firm and we would rather serve one district excellently than promise a state. The ask:
CSC Karnataka designates one district and gives us the working interface to its VLE register. We onboard every centre in the district, and for 90 days we run e-District breadth, the PMFBY 72-hour alert cycle, and the insurance pathway — and we measure one number: incremental VLE income, baselined at day zero, reported to you monthly, ours to stand behind or be judged by.
The commercial structure is deliberately simple, and aligned with that number. A one-time onboarding fee of ₹5,000 per VLE brought onto the platform, borne by CSC Karnataka or the sponsoring programme partner. Thereafter, a monthly success fee of 10% of the incremental income generated through the platform — computed against each centre's day-zero baseline, so it applies only to earnings that did not exist before this work, and settled through the network's existing revenue-share rails as a programme partnership rather than a software procurement. If the number does not move, the success fee does not exist.
GoXero is an applied AI firm in Bengaluru: we build and run accountable automation — systems where AI proposes and a human decides — for businesses and working networks. We are small by design, which is why the proposal is one district and not a state; the model of AI-supported rural agents is proven at national scale internationally, and the corporate programmes now running in Hindi and Telugu show the approach works in India. Kannada, on the CSC network, is open today. We would like to open it with you.
One question we'd ask in the first meeting: is Karnataka on the CAG's state-integration list? That single fact explains much of the income gap — and shapes which services the pilot leads with.
A 45-minute working session — you, us, and the district shortlist. We bring the pilot plan, the measurement design, and the audit-trail architecture; you bring the district and the register. If the 90-day number isn't real, you will have lost nothing and learned something about the gap; if it is real, Karnataka gets the playbook first.
Figures cited are from the Comptroller and Auditor General of India's performance audit of the CSC 2.0 scheme, tabled in Parliament, April 2026. · Contact: nav@goxero.co