One of the significant decisions of the new US government has been the $100 billion Stargate AI initiative. Meanwhile, China is quickly catching up with the US in their bid for AI supremacy by developing cheaper LLMs like QWQ and DeepSeek. These developments are set to transform the global technology landscape.
Like the expansion of the internet and cyber technology, AI’s impact on crime and crime control is inevitable. AI’s advancement propels cybercrime into a new realm, enabling criminals to operate more efficiently with minimal technical skills. Improved text generation tools, realistic image creation, and voice cloning are transforming traditional cyber-crime methods. Deep learning and generative AI are going to transform the modus operandi of criminals.
Story continues below this ad
However, AI also presents an opportunity to improve the capacity of the criminal justice system. Unlike in the West, the Indian system affords less formal discretion to law enforcement and the court systems. In the US, the police can choose not to investigate crimes of lesser significance, and state prosecutors have wide-ranging authority to not pursue and prosecute a case in court. In India, the police are mandated to register all cases, and the courts take up for trial all cases charged by law enforcement authorities. While the system is ideal for a country like India, with its unique socioeconomic diversity, the result has been a slow collapse of the system under its own weight. It is here that AI provides an opportunity for greater efficiency and transparency.
Police in India are burdened by a large number of cases. The introduction of AI to improve the complaint registration process, monitor the progress of investigations, flag the steps to be taken, and assess the quality of the investigation would enhance both efficiency and transparency. AI can also be used in field-level work to analyse data generated in day-to-day operations such as crime location data, patrolling patterns, and routes used by offenders to develop predictive models to guide police operations.
The supervisory burden at the district headquarters can also be greatly reduced through the use of AI for data collation and analysis. With its extensive capacity to process spreadsheets, images, and auditory data, AI can make supervision more efficient and enhance the capacity to deliver justice. At present, a large number of police personnel are used for administrative work, whether related to crime control or personnel management. These officers can be freed for work that could directly benefit the public, such as investigation, law and order patrolling and traffic management.
Story continues below this ad
The government of India has been pursuing the policy of SMART policing, interpreted as Strategic, Meticulous, Adaptable, Reliable, and Transparent policing. AI is a strategic tool as it leverages the large data available to the police and interprets it in a manner much beyond the ability of an individual human. The capacity of AI to process data in various formats, ranging from text to image to voice, ensures that no aspect of a problem is overlooked. Human processing of information may result in inadvertent or willful neglect of crucial evidence during investigation and other decision-making processes. AI can, therefore, improve the rigour and meticulousness of the criminal justice process. The ability of AI to tabulate and present large amounts of data over long periods enhances intra-organisational transparency and makes reliable information available to the public. The goals of SMART policing, which were set by the Prime Minister, could be achieved rapidly with a comprehensive rollout of AI.
The judicial system in India has great legitimacy among the public. However, this well-earned legitimacy is being undermined by a backlog of over 50 million cases. The government and courts have generally pursued a policy of increasing the number of courts and personnel to overcome the colossal backlog. However, in view of the financial burden and the marginal benefits of such measures, it is clear that this strategy is bound to fail. It is here that leveraging technology would be a more feasible mechanism to overcome the fundamental problem of the backlog of cases. To begin with, technology can be used at the trial stage through AI-powered transcription of court proceedings. Management of case files and other documents by automation can make it easier to store, access, and share documents. AI can help detect anomalies, contradictions, or frauds in documents submitted to the court, reducing delays in document authentication and ensuring minimal errors. The technology can also assist in comparisons across decided cases and examining precedence while taking decisions on bail or sentencing. AI would be an asset for judges while conducting research and writing judgments. It could also be used for supervising the workload of lower courts by higher courts. These steps have to be taken after ensuring accuracy, data privacy, and other ethical considerations.
From a policy perspective, the central and state governments could take steps to enhance the efficiency of the criminal justice system through the extensive use of AI. As a first step, a task force should be appointed to study and recommend modalities of the use of AI in public service, especially justice administration. An expedited rollout of AI at police stations, district, and state levels, coupled with comprehensive training of personnel, is essential.
While the two global powers battle it out for AI supremacy, it may be some time before India achieves a place in the sun as a leader in this technology. But, in the meantime, we should use the technology astutely to successfully address administrative and public service problems.
The writer, former DGP and director of Vigilance and Anti-Corruption, Kerala Police, is visiting assistant professor at the University of North Carolina Wilmington