Trend Micro warns of rising AI-Powered Cybersecurity Risks in PH

By TechWatch PH Staff

Artificial intelligence (AI) is reshaping industries and daily life, but its rapid adoption has also opened new doors for cybercriminals. Trend Micro Philippines is urging businesses and organizations to step up their defenses, warning that AI-driven threats are evolving faster than most security systems can keep up.

“People are rushing into AI too fast and are not considering standard security practices to make things secure,” said Morton Swimmer, Principal Threat  Researcher at Trend Micro.

AI’s influence now spans from digital assistants to advanced business automation. But while it enables efficiency, it also enhances the tools available to cybercriminals.

One growing threat is prompt injection attacks such as “Prompt Leakage,” where hackers trick AI models into exposing sensitive information. Deepfakes have also become a leading form of AI-driven scams, with criminals using widely available apps for real-time video manipulation, voice cloning, and other fraud schemes.

Recent global cybersecurity competitions such as Pwn2Own have shown that AI systems, especially those integrated with business infrastructure, can be compromised through overlooked components.

This highlights the importance of securing every layer of the AI ecosystem, from third-party libraries to containerized deployments, alongside adopting proactive strategies like regular security assessments and strong supply chain governance.

Trend Micro emphasized that future-proofing digital assistants such as Microsoft Copilot, Grok, and ChatGPT should be a top priority, as these tools handle large volumes of sensitive data that make them attractive targets. Earlier this year, the firm launched Trend Cybertron, an AI model that can independently analyze environments, automate responses, and coordinate defenses. It also introduced its AI Security Blueprint, offering architecture recommendations for hardening AI systems.

The company said its efforts under the banner #EngineeredToDoGood include publishing annual risk reports, research, and case studies to help organizations adopt practical safeguards against AI-driven threats. This call was further amplified at DECODE 2025, its flagship cybersecurity conference themed “Maximize Momentum,” which focused on sustaining progress in defending against risks brought by emerging technologies.

Since its debut in 2017, DECODE has drawn around 1,000 participants annually, reinforcing its role as a trusted platform for dialogue and upskilling cybersecurity professionals in the country.

Trend Micro also invited the public to access its State of AI Security Report 2025 via www.trendmicro.com.