Identity protected first
Phone numbers, addresses, Aadhaar-like details and personal identity should be removed before anything becomes public.
Makkal Saatchi AI
Turning citizen pain into governance intelligence through safe reporting, AI validation, human review, department routing, action tracking and public-safe transparency.
This is a proposed Makkal Saatchi AI plan, not implemented yet.
Makkal Saatchi is currently a citizen reporting platform. The AI governance workflow below is a future plan that can be piloted with official review, safeguards and department participation.

Clear workflow
The model is designed so every user can understand what happens next. Reports move through validation, review, routing and tracking before becoming public-safe intelligence.
A person safely submits department, district, office, issue, amount, date and optional evidence.
AI checks clarity, location, duplicate signals, evidence strength, risk and routing fit.
A review team approves, rejects, merges, marks sensitive or asks for better detail.
The report is sent to the right department or official dashboard with clean context.
Officials update status, pending cases get reminders, and delays can be escalated.
Only safe public details are shown: district, department, category, status and pattern count.
Repeated public pain becomes measurable intelligence for officers, ministers and civil society.
AI Trust Engine
It does not replace officials. It prepares better information so people can make better decisions faster.
Trust Score
Generated for review, not used as automatic punishment.
Risk Score
Generated for review, not used as automatic punishment.
Duplicate Score
Generated for review, not used as automatic punishment.
Evidence Score
Generated for review, not used as automatic punishment.
Pattern Match
Generated for review, not used as automatic punishment.
Routing Confidence
Generated for review, not used as automatic punishment.
Priority Level
Generated for review, not used as automatic punishment.
Phone numbers, addresses, Aadhaar-like details and personal identity should be removed before anything becomes public.
The engine does not punish anyone or declare guilt. It prepares reports for human review and official action.
One anonymous report may be hard to prove, but many similar reports from the same place can reveal a serious hotspot.
Why repeated reports matter
The goal is not to blindly believe every complaint. The goal is to identify repeated public pain points faster and help honest officers respond with better context.
20 reports about an EB office name-transfer bribe
30 reports about registration-office delay or payment demand
Many e-Sevai complaints repeating in the same area
Several patta-related bribe demands from one revenue office
Who this helps
A safe way to speak without fear, with clearer follow-up questions before submission.
Clean, prioritized reports instead of scattered noise and vague public complaints.
Tracking, escalation, audit logs, action history and status visibility.
Hotspots, SLA delays, high-risk reports and early warning signals before issues become crises.
Practical pilot
This does not need to start big. A focused pilot can test safety, report quality, official workflow and action tracking before wider expansion.
If one citizen speaks, it is a complaint. If thousands speak safely, it becomes governance intelligence.
Public transparency should protect citizens, not endanger them. AI should help governance listen better, not become governance by itself.