AI in Policing: Technology with Responsibility

The decision of Haryana Police to establish an AI Task Force, with a particular focus on cybercrime and AI-assisted financial trail analysis, is a welcome and timely initiative.

Cybercrime has fundamentally altered the investigative environment. A single case may involve thousands of transactions, multiple bank accounts, digital identities, devices and jurisdictions. In such situations, technology can help investigators identify patterns, connections and anomalies that may be difficult to detect through conventional analysis alone.

The proposed use of AI for financial trail analysis is particularly significant. Used well, it can help investigators connect seemingly unrelated transactions, trace movement of funds, prioritise leads and make better use of limited investigative time. Similar applications in citizen grievance redressal and other police processes can also improve responsiveness and efficiency.

However, as AI becomes more deeply embedded in policing, I believe three principles deserve equal attention.

First, AI should augment professional judgment, not replace it. An algorithm may identify a correlation or flag a transaction, but the decision about its investigative significance must remain with a trained officer.

Second, capability must develop alongside safeguards. Data quality, privacy, cybersecurity, auditability and protection against algorithmic bias will be as important as the sophistication of the technology itself. In policing, where technological outputs can affect individual rights and reputations, accountability cannot be an afterthought.

Third, collaboration will be critical. Cybercrime rarely respects state boundaries. If different police forces are developing AI capabilities independently, there may be considerable value in creating common standards, interoperable tools, shared learning and mechanisms for exchanging successful practices. Institutions such as the Indian Cyber Crime Coordination Centre (I4C) could have an important role in facilitating such convergence.

Ultimately, the measure of “smart policing” should not be how much AI is deployed, but whether technology helps the police become more effective, more responsive, and equally, more fair and accountable.

Haryana Police's initiative is therefore encouraging—not simply because it introduces a new technology, but because it provides an opportunity to think seriously about how technology and human judgment can complement each other in the service of better policing.