In Starkville, a small town in the United States, police are able to access feeds from all private and public cameras using hardware and software developed by a company called Fusus.
The AI-enabled system funnels live feeds from usually siloed cameras into one central location, and scans for people wearing certain clothes, carrying a particular bag, or traveling in a certain vehicle.
The hardware is called SmartCORE and adds AI to ordinary surveillance cameras, then allowing the police to detect different objects, vehicles, clothing and people.
In short, Fusus turns dumb cameras into smart ones, or a brain is put into every camera connected into the system. Such AI-powered systems are rapidly springing up across small town America and major cities alike. At the recent International Association of Chiefs of Police (IACP) Conference, Fusus was the centre of attraction.
The advantages were brought out by Starkville police chief Mark Ballard: “If we have a guy that walks into a crowd with a gun and we pick him up on camera, our officers are on top of him in a matter of a minute, seconds, really, where back before the cameras we may would have to wait till the incident actually happens or somebody else sees it”.
Despite the benefits, concerns have arisen regarding the widespread deployment of comprehensive surveillance systems, especially in India. In India, police accountability primarily aligns with senior ranks, a system originating from colonial influences.
During the introduction of the Police Act, the British chose a system, akin to the French Prefecture model, where magistrates and police superintendents were accountable only to their superiors in the hierarchy. This vertical accountability was further solidified by the army’s significant influence on the police’s structure and operations, including multiple ranks and formal reporting from lower to higher ranks.
That the army and the police are different was brought out crisply by a Director of the National Police Academy during the 1980s:
For the Army, Land is the objective and Man the hinderance; For the Police, Man is the objective and Land the hinderance.
AI-driven policing mimics human eyes through smart cameras, enabling real-time monitoring without physical presence. This innovation overcomes the obstacle represented by ‘land’. In order to reach the ‘man’, horizontal accountability to citizens has to be built into the structure and operating systems of the police.
For a shift towards horizontal accountability, two significant changes in police operating systems are proposed.
- Firstly, sharing real-time auditable access to camera feed logs and surveillance analytics with citizens.
- Secondly, involving local residents in decisions concerning street safety and traffic management in partnership with the local police stations.
As part of the move towards horizontal accountability, some modifications in the structure of the police would go alongside changes in the operating systems. Some key changes are:
- Redistribution of power from districts to police stations by reducing hierarchical levels between police commissioners and stations.
- Assigning senior officers like DySPs/SPs/DIGs as chiefs of police stations.
- Consolidating police stations and creating specialized verticals for police functions.
AI-driven policing holds significant potential for enhancing law enforcement in India, particularly in crime prevention. Decentralizing authority to police stations and involving residents in decision-making would make citizens take a more active interest in policing that matters to them. This would ignite horizontal accountability of the police and serve as the foundation for fundamental reforms in the police force.