Singapore, one of Asia's most technologically advanced financial hubs, has faced a new wave of cyber threats where artificial intelligence has become the primary weapon of attackers. My analysis of the latest market data reveals a troubling trend: 79% of local organizations have been subjected to at least one AI-powered attack over the past year. This is not just statistics—it is a signal of a paradigm shift in digital security.
The Three Pillars of AI Attacks
Three dominant vectors stand out in the structure of incidents. The most widespread are phishing campaigns and impersonation based on generative models—46% of companies have suffered from these. Slightly less critical, but no less dangerous, are "prompt injections" and targeted attacks on the AI tools themselves—41%. Finally, deepfakes and voice cloning are recorded in 39% of cases. These figures show that attackers have moved from mass spam to targeted, highly personalized operations, where AI amplifies social engineering to an unprecedented level.
The Governance Gap
The key issue I see in the data is the gap between adoption and control. 97% of respondents are already actively using or piloting AI solutions, but only 49% of them have mechanisms for access control and verifying the outputs of these systems. This creates a paradox: companies voluntarily open up new attack surfaces without ensuring basic security hygiene. The survey, which covered 400 cybersecurity decision-makers, underscores the systemic nature of the problem.
My expert assessment: Singapore is an indicator of a global trend. If AI risk management is lagging here, despite high digital maturity, then the situation could be catastrophic in less prepared markets. Businesses urgently need to revisit their policies: adopting AI without robust governance is not innovation—it is a ticking time bomb.