PETALING JAYA (ANN/THE STAR) – Malaysia is among nations aiming to use artificial intelligence (AI) in investigating commercial crimes.
However, the Artificial Intelligence Society Malaysia has urged the police force to first guarantee that the men in blue receive thorough training.
Its president Dr Azree Shahrel Ahmad Nazri said police officers need to understand how AI tools function, interpret AI-generated insights correctly and ensure their investigations maintain human oversight.
Early this month, the Bukit Aman Commercial Crime Investigation Department (CCID) said the use of AI would ensure investigating officers could optimise their time when managing cases.
Azree Shahrel suggested that Bukit Aman consider collaborations with Malaysian universities and AI experts to develop tailored training modules for CCID officers and cybercrime units.
The police, he said, should begin AI implementation by focusing on commercial crimes, cyberfraud and money laundering investigations.
Azree Shahrel, the head of laboratory at the Institute of Mathematical Research at Universiti Putra Malaysia, has suggested integrating AI-powered chatbots and automated scam alerts into banking systems to warn users of suspicious transactions.
“This proactive approach not only helps in crime prevention but also educates the public about ongoing threats,” he said.
Additionally, he said the police could collaborate with local fintech companies, cybersecurity firms, and AI research institutions to develop AI-driven crime prevention systems.
“This will allow Malaysia to stay ahead of evolving crime tactics while ensuring law enforcement agencies are equipped with the latest technology,” he added.
Azree Shahrel emphasised that AI-driven fraud detection systems employed by Malaysian banks have successfully identified suspicious transactions related to mule accounts, often used by scammers to launder stolen money.
“AI can track money trails across multiple accounts, something that would take human investigators weeks or months to complete. Additionally, AI-powered natural language processing tools can analyse scammer conversations on messaging apps like WhatsApp and Telegram, where many fraudulent investment schemes are promoted. In cybercrime investigations, AI can assist in identifying fake websites, detecting phishing attempts and analysing large amounts of digital evidence,” he said.
Azree Shahrel also said the university’s Institute of Mathematical Research and Faculty of Computer Science and Information Technology have developed PutraGPT, which aids in fraud detection, identifying suspicious transactions and tracking complex financial trails.
“In cybercrime investigations, PutraGPT helps to identify fake websites, detect phishing attempts, and process vast amounts of digital evidence,” he said, citing the recent fake e-commerce websites that were detected by the system.
“By automating these processes, PutraGPT can improve overall cybersecurity, offering a faster and more accurate response to evolving financial threats.”
