AI system
- System
-
Israeli facial recognition software used by Delhi Police (not named in the reports)
- Model
-
Not assessed
- Release
-
Not assessed
- Provider
-
Not assessed
- Role
-
Not assessed
- Configuration
-
Not assessed
- Inputs and outputs
-
Not assessed
- Autonomy
-
Not assessed
- Workflow
-
Dense CCTV coverage (about 500 cameras across five areas) from which a usable frame of the suspect's face could be taken. A police facial recognition database that already held the suspect's photograph from a prior arrest at Kotwali police station.
- Human oversight
-
Not assessed
- Relation to AI
-
By DCP (North) Raja Banthia's account, teams took a screengrab of a man seen following the victim. The Israeli software matched it to a prior record from Kotwali police station with a similarity score of 32.09 percent. That gave them Ali's name and his Daryaganj address. A score that low is a lead, not an identification. Police confirmed it by laying a trap and interrogating Ali. The auto-rickshaw trace, the reconnaissance evidence and Samir's arrest came from conventional work. The reports name neither the software nor its vendor.
- AI contribution
-
By DCP (North) Raja Banthia's account, teams took a screengrab of a man seen following the victim. The Israeli software matched it to a prior record from Kotwali police station with a similarity score of 32.09 percent. That gave them Ali's name and his Daryaganj address. A score that low is a lead, not an identification. Police confirmed it by laying a trap and interrogating Ali. The auto-rickshaw trace, the reconnaissance evidence and Samir's arrest came from conventional work. The reports name neither the software nor its vendor.
Benefits
- Outcome identifiers
-
Not assessed
- Benefit
-
two arrested within about a day and Rs 79.5 lakh of the Rs 80 lakh recovered
- Category
-
Reduced Violence
- Status
-
Not assessed
- Beneficiary groups
-
People. Organizations. Society.
- Beneficiaries
-
Not assessed
- Reach
-
Not assessed
- Distribution and access
-
Not assessed
- Duration
-
Not assessed
- Expected?
-
Yes
- Costs and harms
-
Not assessed
Measurement
- Studies
-
Not assessed
- Study design
-
Not assessed
- Measurement
-
Not assessed
- Setting and sample
-
Not assessed
- Measures
-
two arrested within about a day and Rs 79.5 lakh of the Rs 80 lakh recovered
- Comparison
-
Not assessed
- Results
-
two arrested within about a day and Rs 79.5 lakh of the Rs 80 lakh recovered
- Additional value from AI
-
Not assessed
- Limitations
-
Not assessed
- Denominators
-
Not assessed
- Materials for replication
-
Not assessed
Money
- Kind of value
-
stolen cash recovered
- Amount
-
Rs 79.5 lakh
- Currency and prices
-
Not assessed
- Period and unit
-
Not assessed
- Realization
-
Not assessed
- Recipients and payers
-
the robbed cash courier's firm, R K Enterprises (recovered by Delhi Police from the accused's residence)
- Gross or net
-
Not assessed
- Included costs
-
Not assessed
- Comparison
-
Not assessed
- Method
-
Not assessed
- Return and payback
-
Not assessed
People and places
- Participants
-
Operator:
Delhi Police, North district. - Jurisdictions
-
Benefit: Delhi,
India. Operator: India. - Ownership and control
-
Not assessed
- Dependencies
-
Not assessed
- Cross-border connections
-
Not assessed
- Geopolitical implications
-
Not assessed
Follow-up
- Conditions for replication
-
A low-similarity match (32.09 percent) is only a lead and must be confirmed by conventional investigation before arrest. The suspect must already be in the database; a first-time offender would not have been matched.
- Replication attempts
-
Not assessed
- Later findings
-
No chargesheet, trial or other development after the 19-20 March 2025 arrest reports was found; the reports say only that both accused were booked.
- Availability
-
Not assessed
- Institutional decision
-
Not assessed
- Grounds
-
Not assessed
- Action and result
-
Not assessed
- Success criterion
-
Not assessed
Evidence
- Source numbers
-
AIBD-0126-source-0001–0005
- Citations
-
Numbered references below.
- Language
-
Not assessed
- Source relationships
-
Independent
- Original evidence
-
Not assessed
- Access and versions
-
Not assessed
- Supported findings
-
Not assessed
- Evidence links
-
Not assessed
- Evidence strength
-
Not assessed
- Missing evidence
-
Not assessed
Grades
- Assessed benefits
-
People. Organizations. Society.
- Depth
-
People: 2. Organizations: 2. Society: 1.
- Reasons
-
People: Ajmal Bhai Ganesh and his employer R K Enterprises recovered Rs 79.5 lakh of the Rs 80 lakh taken at gunpoint. Police described that as the entire amount. A large material loss reversed within two days. Organizations: Delhi Police's North district cleared an armed robbery whose video had gone viral within about 24 hours. Both accused are in judicial custody under the BNS and Arms Act and the pistol recovered. Society: Two men who had surveilled cash couriers and used a firearm were taken off the street. The reports give no evidence of further robberies prevented, so the public benefit is small.
- Observed or intended
-
Not assessed
- Rubric
-
Not assessed
- Assessor and date
-
Not assessed
- Alternative assessments
-
Not assessed
- Fill
-
Independently reported
Entry details
- Kind
-
Not assessed
- Related entries
-
Not assessed
- Field
-
Policing / robbery investigation. CCTV-based suspect identification in an armed robbery case.
- Purpose
-
The software was deployed to identify people in CCTV footage by matching faces against police databases. Naming a robbery suspect from a frame is that function.
- Dates
-
Date basis unassessed: 2025 (precision unassessed). Updated: 21 October 2026.
- Stage
-
Not assessed
- Version
-
SHA-256: 583679466697…
References
- Other identifiers
-
oecd_aim: 2025-03-19-2b12.
- Category mappings
-
Not assessed
- Brief
-
Not assessed
- Revisions
-
5 revisions. Latest: 21 October 2026.
Sources
-
News report Independent author
The Times of India (2025). Chandni Chowk heist: Israeli face recognition software ends robbers' run within 24 hrs. The Times of India, Delhi, 19 March 2025.
A breakthrough came when Mohammad Ali's face was captured on CCTV near the crime scene. Using Israeli facial recognition software, our teams matched it with a prior criminal record
-
News report Independent author
The Indian Express (2025). CCTV footage, face recognition tech: How Delhi cops nabbed two men who stole Rs 80 lakh from trader. The Indian Express, 20 March 2025.
As police analysed footage from around 500 CCTV cameras, they identified Ali in the clips — as he followed Ajmal — through a facial recognition system.
-
News report Independent author
Press Trust of India via Hindustan Times (2025). Israeli facial recognition tech helps Delhi Police crack ₹80 lakh robbery case; 2 held. Hindustan Times, 19 March 2025.
An Israeli facial recognition system has helped Delhi Police solve a case of robbery at gunpoint, arrest two accused persons, and recover the entire stolen amount of ₹80 lakh
-
News report Independent author
Press Trust of India via Rediff.com (2025). Israeli tech helps Delhi police crack Rs 80 lakh robbery; 2 held. Rediff.com, 19 March 2025.
The similarity score was 32.09 percent. We got to know that the accused lived in Daryaganj, where teams were already working to trace the Scooty used in the crime
-
News report Independent author
The New Indian Express (2025). Israeli facial recognition tech helps Delhi police crack Rs 80 lakh robbery case, two held. The New Indian Express, 20 March 2025.
Facial recognition technology identified one suspect as Mohammad Ali, who had been arrested previously in a similar case, Deputy Commissioner of Police (North) Raja Banthia said.