Maha Kumbh, the world's largest human religious gathering, is making use of advanced technology like Artificial Intelligence to manage crowds and prevent stampedes.
Organisers of Maha Kumbh 2025 predict that this year's Kumbh Mela will attract up to 400 million pilgrims. According to official estimates, some six million people took the holy dip on the first morning of the festival. The massive religious gathering, an age-old tradition, has a history of deadly crowds and stampedes. To avoid such crowd crushes, this year, the organisers have put AI to use.
Also read | Second Amrit Snan at Maha Kumbh set for Jan 29, preparations underway
Talking to AFP, Amit Kumar, a senior police officer heading tech operations in the festival said: "We want everyone to go back home happily after having fulfilled their spiritual duties," and that to ensure that "AI is helping us avoid reaching that critical mass in sensitive places."
Stampedes and Maha Kumbh
In 1954, in one of the highest single-day tolls in a crowd-related disaster globally, more than 400 were trampled to death or died by drowning at the Kumbh mela.
In 2013, the last time Maha Kumbh was held in Prayagraj (then called Allahabad) 36 people were crushed to death.
Also read | Musk, Putin and John Cena at Maha Kumbh? AI video takes over internet, leaves netizens amused
However, this time, the organisers say that the technology deployed will help them gather accurate estimates of the crowd sizes and allow them to be better prepared for any potential trouble.
To aid in this, as per the police, around 300 cameras have been installed across the festival site. In addition, overhead drones have been deployed, all of which are monitored by a small army of police officers and technicians located in a glass-panelled command-and-control room near the spiritual centre of the festival.
Also read | Maha Kumbh Mela 2025 sees record attendance, 70 million devotees take a dip in sacred waters
This footage, as per Kumar, is fed into an AI algorithm that provides the handlers an overall estimate of the crowd, which is then cross-checked against data from railways and bus operators.
"We are using AI to track people flow, crowd density at various inlets, adding them up and then interpolating from there," he added.
The AI-based system sounds an alarm if there are sections of crowds so concentrated that they pose a safety threat.
(With inputs from agencies)