A Crisis of Alarming Scale
The statistics surrounding missing persons in India are staggering. According to the National Crime Records Bureau (NCRB), hundreds of thousands of individuals are reported missing each year, with a significant number remaining untraced. Between 2019
and 2021 alone, over 13 lakh girls and women were reported missing. The numbers for children are equally distressing; in 2020, reports were filed for over 59,000 missing children. For police forces, who are often overworked and under-resourced, sifting through leads, photos, and endless hours of footage is a monumental task. Every minute that passes diminishes the chances of a safe return, making speed and efficiency critical components of any investigation.
How AI Enters the Picture
Artificial intelligence offers a solution to the challenge of scale and speed. Unlike human investigators, AI systems can process vast amounts of information in minutes. The primary tool in this new arsenal is facial recognition. These algorithms can scan thousands of hours of CCTV footage from public spaces like railway stations, bus terminals, and markets, searching for a match. They can also compare a missing person's photo against massive databases, including those of government childcare institutions and NGOs. In one early trial, Delhi Police used the technology to scan 45,000 photos of children in institutional homes, identifying nearly 3,000 missing children in just four days.
Beyond Simple Matching
Modern AI offers more than just straightforward photo matching. Generative AI is being used to enhance old, faded, or low-quality photographs of missing individuals, creating clearer images for circulation. Some police departments, like those in Rajasthan, are collaborating with AI artists to use age-progression technology, creating an estimation of what a child who went missing years ago might look like today. These digitally enhanced or aged images provide a much more effective tool for public identification. In one case, Delhi police even used AI to reconstruct the face of a deceased victim who was beyond recognition, which led to his identification and the eventual arrest of his killer.
Connecting the Dots
Another powerful application of AI is its ability to analyse data to find patterns that humans might miss. An AI can scan a missing person’s social media activity for location tags or signs of emotional distress. It can also help connect disparate pieces of information across jurisdictions. For instance, a person admitted to a hospital in one state could be the same individual reported missing in another. Without an integrated system, these dots are rarely connected. AI-powered platforms can automatically flag such potential matches, providing investigators with crucial, actionable leads. Portals like Khoji.in are already being used by some police forces to centralise this data and leverage AI for matches.
A Tool, Not a Magic Wand
Despite its immense potential, AI is not a perfect solution. The use of this technology in law enforcement raises significant ethical questions. In a country without robust data privacy laws, the large-scale collection and analysis of personal data is a major concern. Furthermore, AI algorithms are only as good as the data they are trained on. If the training data contains biases related to gender, caste, or religion, the AI can replicate and even amplify those biases in its results. Experts stress the need for strong regulatory frameworks and human oversight to ensure the technology is used responsibly and that accountability is clear when errors occur.
