UPSC Mains 2026 GS4 Q8: model answer
Ravi is a senior police officer with vast experience in riot control and cyber-policing. Since one year, he has been the Superintendent of Police (SP) of a district with a history of frequent rioting. Last year, Ravi had sought installation of an AI enabled software for predictive policing. This system has been operational for approximately six months. This new system employs advanced algorithms for capturing the biometric data of persons in a crowd and swiftly relating it to a data library. This has enabled the police to identify the persons involved in various crimes. The system has identified an immigrant and low-income neighbourhood as a centre for gang violence and drug trafficking. Aided by this AI analysis, the local police has focused its patrolling, preventive detentions and establishing checkposts. Consequently, public order and law enforcement has visibly improved. Last week, some community leaders, civil rights lawyers and human rights activists visited Ravi's office. They submitted a memorandum that the new system is faulty as it is based on incorrect historical data caused by social biases and discriminatory policing. The memorandum also alleges that the increased surveillance has created a climate of tension amongst residents. This feeling is aggravated by the fact that the residents are not aware of the data noted against their names.
(a) What are the ethical issues including biases involved in the use of AI in data-driven policing?
(b) Place yourself in Ravi's role and discuss the alternatives available. Justify the action that optimises compliance with ethics.
What the question wants
(a) Ethical issues of AI policing: biased data, discrimination, privacy, transparency, due process, accountability. (b) Options for Ravi, with the best one justified.
How to structure it
(a) Issues list with the feedback-loop point. (b) Options: continue, scrap, or audit-and-reform; choose the last and justify.
- 1Part (a): Issues list with the feedback-loop point
- 2Part (b): Options: continue, scrap, or audit-and-reform; choose the last and justify
Model answer
(a) Ethical issues
- 1Past over-policing of an area
- 2More records from that area
- 3Algorithm flags it as high-risk
- 4More patrols and detentions
- 5More records: the loop repeats
Repeats: the last stage feeds back into the first
- Biased training data: if past policing targeted poor and immigrant areas, the algorithm learns that bias and sends more police there, generating more arrests: a self-reinforcing feedback loop.
- Discrimination: profiling by place and class violates Article 14 and the principle of equal protection.
- Privacy: mass biometric capture in crowds without a clear legal basis fails the Puttaswamy (2017) tests of legality, necessity and proportionality.
- Transparency and due process: residents do not know what data is held against them and cannot correct it. Preventive detentions based on a black box deny fairness.
- Accountability: who answers for a wrong match, the vendor, the software or the SP?
- Trust: improved order on paper may hide fear and alienation, which harms policing in the long run.
(b) Options before Ravi
- Continue as is: visible results, but ignores rights and bias; risks court challenge.
- Scrap the system: answers the complaint but discards a useful tool.
- Pause intrusive uses, audit and reform: the best option.
Recommended action
- Suspend preventive detentions based only on AI flags; require human verification and independent evidence.
- Order an independent bias audit of data and outcomes, with experts and civil society.
- Limit biometric capture to specific investigations, with data retention rules in line with the DPDP Act, 2023.
- Tell residents what data is held and create a grievance and correction process.
- Community policing meetings in the neighbourhood to rebuild trust.
This keeps technology as an aid, not a judge, and meets both effectiveness and fairness. Policing by consent remains the ethical and practical goal.
Keywords, cases and data to drop in
algorithmic bias feedback loop proportionality due process bias audit community policing
Where most answers lose marks
- Rejecting technology outright
- No concrete safeguards
Fresh examples from current affairs
Govt outlines regulatory steps against AI deepfakes
In a Lok Sabha reply released on 6 August 2026, MeitY set out how Indian law tackles AI deepfakes. It covers the IT Act, 2000, the Bharatiya Nyaya Sanhita, 2023 and the IT Rules, 2021. The 10 February 2026 amendment requires labels and traceable metadata on AI-generated content and cuts some takedown timelines from 36 hours to 3 hours.
UPSC angle: Mains: Regulating deepfakes.
6 Aug 2026
AI traces 100 years of Kodaikanal Sun records
A team led by ARIES, an autonomous institute of DST, used machine learning (U-Net) to read hand-drawn sun charts from the Kodaikanal Solar Observatory. It traced bright magnetic patches called plages from 1916 to 2007, across nine solar cycles. The results match Kodaikanal's Ca II K images and can fill gaps in long-term solar data.
UPSC angle: Prelims: observatories.
1 Jul 2026
PM at G7 session on safe and rapid AI rollout
At the France-hosted G7 Summit, the PM spoke on 17 June 2026 on ensuring a safe, rapid and efficient AI rollout. He stressed human-centric design under India's MANAV vision from the AI Impact Summit, and urged safe-by-design systems, common standards, action on deepfakes and Global South access.
UPSC angle: Prelims: G7 members.
17 Jun 2026
PM’s France visit: Innovation Roadmap 2030 and AI working group
On the Prime Minister's France visit outcomes released on 14 June 2026, India and France adopted the Innovation Roadmap 2030 and created a Joint AI Working Group on AI governance. The package includes a National Centre of Excellence for skilling in aeronautics at NSTI Kanpur, wider UPI use in France, and ISRO–CNES cooperation on microgravity and human spaceflight.
UPSC angle: Prelims: Horizon 2047, Station F.
14 Jun 2026Learn the concept
Question text as printed in the official UPSC question paper: https://www.upsc.gov.in/sites/default/files/QP-CSM-26-010926-GENERAL-STUDIES-PAPER-IV.pdf The model answer is original Battle IAS work; use it to learn structure, not to memorise.