Pakistan: FIA deploys AI system to track most wanted criminals in real time

Dubai: In a major step toward modernising law enforcement, Pakistan’s Federal Investigation Agency (FIA) has introduced an artificial intelligence (AI)-powered system capable of tracking wanted criminals by generating real-time visual updates of their appearance.

The initiative, described as part of the agency’s broader digital transformation drive, will enable authorities to identify suspects even if they have altered their physical features over time.

FIA Director General Dr Usman Anwar said the upgraded system would make it “impossible” for criminals to evade detection simply by changing their looks, Dawn news reported.

At the heart of the reform is a revamped digital “red book”, a database of the country’s most wanted individuals, now enhanced with AI tools that can recreate current facial appearances using older photographs. The system can simulate changes such as ageing, hair loss, or facial hair, significantly improving identification accuracy.

Online ‘red book’

The online 'red book' will be accessible to relevant FIA officials and the public, and will include comprehensive profiles of suspects. These profiles feature details such as family members, known associates, CNIC and passport information, phone numbers, distinguishing marks, bank account data, FIR records, and the status of court proceedings.

Get updated faster and for FREE: Download the Gulf News app now - simply click here.

Anti-human trafficking

In a significant boost to anti-human trafficking efforts, the database will also document the modus operandi, operational areas, routes, and last known locations of traffickers. According to the FIA, at least 143 human smugglers currently wanted in various cases have been listed.

Dr Anwar emphasised that embracing modern technology is central to the FIA’s reform agenda, aimed at replacing fragmented manual processes with integrated, data-driven systems. “Digitalisation enhances transparency, efficiency, and oversight, while strengthening coordination between departments,” he said.

He added that ongoing reforms include the digitisation of investigations, inspections, accountability mechanisms, and human resource processes, steps designed to institutionalise e-governance and modern policing practices.

Accountability

The FIA chief also noted that internal accountability systems have already been digitised, including fact-finding inquiries and departmental proceedings, with further improvements planned. Meanwhile, the Centralised Immigration Monitoring and Control Centre at FIA headquarters is being upgraded and relocated to enhance operational capacity.

Addressing concerns over migrant boat tragedies, Dr Anwar said strict measures are being implemented to curb illegal migration and human smuggling. He highlighted strengthened legal frameworks and an ongoing crackdown against traffickers, adding that negligent officials are being held accountable.

He also urged young people seeking employment opportunities abroad to pursue legal channels and avoid risking their lives through illegal means.

Financial fraud

Separately, former FIA director general Dr Sanaullah Abbasi highlighted the growing role of AI in combating financial fraud, noting that scams have surged globally due to the expansion of digital transactions and e-commerce.

He explained that traditional fraud detection methods are increasingly ineffective against sophisticated, technology-driven schemes. In contrast, AI systems can analyse vast volumes of transactional and behavioural data in real time, identifying anomalies and evolving alongside emerging fraud patterns, The News reported.

Deep learning

“AI enables predictive analytics, allowing organisations to detect and prevent fraud before it occurs,” Abbasi said, adding that technologies such as machine learning, deep learning, and natural language processing are transforming how businesses safeguard their operations.

He pointed out that modern fraud schemes often involve automation, synthetic identities, phishing, and coordinated attacks, making manual detection nearly impossible. AI-driven systems, however, can integrate multiple data sources and uncover complex fraud networks through advanced techniques like graph analysis.

‘Black box’

Despite its advantages, Abbasi acknowledged challenges such as data privacy concerns, imbalanced data sets, and the “black box” nature of some AI models. He stressed the need for hybrid approaches combining rule-based systems with advanced AI, alongside stronger encryption and explainable AI frameworks.

He concluded that as fraud becomes more sophisticated, AI offers a scalable and adaptive solution, enabling organisations to stay ahead of evolving threats while improving trust and compliance in an increasingly digital economy.