Despite the rapid advancements in technology and cybersecurity measures, the human factor continues to pose significant vulnerabilities in cyberattacks. At the Protect Insights forum held last week, cybersecurity firm Proofpoint Inc emphasized the necessity of a human-centric approach to combat threats in the era of artificial intelligence (AI). “Humans are still humans. They are still being careless. Of 5,735 [cyberattack] cases reported, 90% involve humans [making mistakes], and this only covers reported cases,” stated George Lee, senior vice-president for the Asia-Pacific and Japan regions at Proofpoint, during a media roundtable at the Four Seasons Hotel on March 3.
Lee highlighted alarming trends from Malaysia’s Cyber Security Response Centre, noting a more than 20% year-on-year increase in cyberattacks as of September 2025, with nearly half attributed to phishing attempts. Phishing, which tricks users into disclosing private information or credentials, remains a prevalent method for attackers seeking to gain access to company systems and sensitive data.
He explained that the rise of agentic AI—systems capable of autonomously executing tasks—has broadened potential attack surfaces, with phishing emails still being the most common attack vector. “Language itself was a deterrent against phishing emails, because when a scammer tries to emulate [a foreign language], it typically wouldn’t work. But with the resurgence of AI today, it can emulate a language to an almost human-like interaction, which has been a game-changer for the bad guys,” Lee remarked.
Moreover, some cyberattacks are now exploiting AI tools used to summarize emails by embedding hidden instructions that AI systems can recognize, but human users cannot detect. “AI innovation must be matched with AI security. By adopting a human-centric cybersecurity strategy, we hope to help organizations in Malaysia accelerate innovations securely and confidently, strengthening digital trust as the foundation of the AI-driven future,” he added.
In another development, Digital Nasional Bhd (DNB), in collaboration with Ericsson, has introduced a new capability within its 5G network to detect illegal transmitters used for sending scam messages to mobile users. Announced at the 2026 Mobile World Congress in Barcelona on March 3, the initiative is touted as the world’s first trial of a Radio Access Network (RAN) and network management enhancement designed to unveil false base stations within a commercial 5G network.
This capability has been deployed across DNB’s national 5G network using Ericsson’s network software and the Ericsson Security Manager cybersecurity platform. It aims to identify fake transmitters utilized by scammers to impersonate legitimate mobile base stations, allowing them to send malicious messages aimed at stealing personal information.
According to the GSMA Asean Consumer Scam Report 2025, the proportion of consumers who report having been scammed has increased significantly, from 31% to 43%. In Malaysia, authorities blocked 2.1 billion suspicious calls and unsolicited messages from 2022 to August 2025. “Our foremost priority has always been to deliver a secure and trusted national 5G network for the people of Malaysia,” DNB CEO Datuk Azman Ismail stated.
In a separate initiative to facilitate AI adoption, V Gallant Sdn Bhd has launched an AI-focused co-working and computing hub at KL Eco City Mall. This facility, equipped with Blackwell Graphics Processing Units (GPUs) from Nvidia, offers on-demand access to high-performance computing resources for companies and start-ups without requiring substantial upfront investments. V Gallant describes the GPU Lounge as Malaysia’s first facility of its kind, designed to make AI development more accessible.
“We are your AI transformer. Our focus is not only on enabling access to technology but on working closely with organizations as a consultative partner,” said Chief Operating Officer Chan Wai Mun. Pricing for access starts at RM7 per hour for a range of computing resources, scaling up to RM5,125 per month, with higher-tier configurations available for more demanding AI tasks.
In the realm of venture capital, Gobi Partners has invested in Cortical Labs, a Melbourne-based company developing biological intelligence systems that integrate living human neurons with silicon hardware. This funding round also includes participation from global investors and aligns with Malaysia’s Dana Impak initiative to bolster capabilities in forward-looking technologies.
Dr. Hon Weng Chong, founder and CEO of Cortical Labs, emphasized that their work addresses rising constraints in AI related to energy and data efficiency. “Just as the world has woken up to the economic benefits of AI, we are now hitting the wall with energy and data efficiency,” he said.
In mobility, Bolt has expanded its ride-hailing platform to include licensed taxis in the Klang Valley, integrating traditional taxi services into its existing operations. This move aims to enhance ride availability and reduce wait times for passengers.
In 2025, Malaysia faced over 18.8 million web-based cyber threats, averaging over 51,000 attempted malicious incidents daily, as reported by the Kaspersky Security Network (KSN). This data underscores the ongoing challenges businesses face in maintaining cybersecurity, particularly as the country seeks to expand its digital economy.
As Malaysia’s digital landscape evolves, the emphasis on human-centric cybersecurity, innovative technology deployments, and strategic partnerships will be crucial in navigating the complexities of an increasingly interconnected world.
See also
OpenAI Launches Codex Security for Context-Aware Vulnerability Detection, Cutting Noise by 84%
Anthropic’s Claims of AI-Driven Cyberattacks Raise Industry Skepticism
Anthropic Reports AI-Driven Cyberattack Linked to Chinese Espionage
Quantum Computing Threatens Current Cryptography, Experts Seek Solutions
Anthropic’s Claude AI exploited in significant cyber-espionage operation
