Insider Spotlight
Cybersecurity firm Trend Micro Philippines is calling for stronger defenses, warning that organizations must adopt proactive strategies to secure AI systems. This includes enhancing risk governance and developing security patterns that keep pace with evolving threats.
“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.
Vulnerabilities posed by AI-based solutions
AI is proving to be a powerful business enabler—but it is also supercharging cybercriminal capabilities. One key risk is indirect prompt injections, where malicious content leads to the leakage of user-uploaded data. A common technique is Prompt Leakage (Pleak), which can expose sensitive instructions embedded within AI systems.
Another major concern is the rise of deepfakes, which have become the leading AI-driven scam. Criminals are increasingly exploiting publicly available—and sometimes free—applications to create manipulated videos, voices, and images in real time.
The sophistication and scale of AI-driven cyberattacks are rapidly evolving. Incidents at global competitions such as Pwn2Own show how AI systems integrated into business infrastructure can be compromised through overlooked or insecure components. This highlights the need to secure every layer of the AI ecosystem, from third-party libraries to containerized deployments.
Rushing development, rising risks
In the race to roll out AI applications, many developers are neglecting rigorous testing and audits. Unlike traditional software, custom AI systems often undergo less scrutiny, making them more vulnerable to attacks.
This lack of due diligence can leave systems prematurely exposed to the internet and external threat actors. Trend Micro emphasizes that regular security assessments and robust supply chain governance are critical to mitigating these risks.
Future-proofing digital assistants
AI-driven services such as Microsoft Copilot, Grok, and ChatGPT are reshaping how organizations interact with technology, automating workflows and handling sensitive data. As these tools evolve, they are becoming attractive targets for attackers.
Recognizing this, Trend Micro recently launched Trend Cybertron, an AI-powered agent designed to analyze environments, automate incident response, and act on cyberthreats. The company also unveiled its AI Security Blueprint, a set of architectural guidelines to harden AI systems and bridge the gap between awareness and practical security steps.
Through research initiatives and open sharing of intelligence, Trend Micro aims to strengthen global cybersecurity defenses. Its Annual Risk Report and case studies are made publicly available to help organizations of all sizes identify risks, apply mitigation techniques, and prepare for AI-driven threats.
DECODE 2025: Maximizing momentum
In a world shaped by AI, machine learning, and quantum computing, businesses cannot afford to lag in cybersecurity. Trend Micro’s DECODE 2025: Maximize Momentum conference underscored the importance of using past innovations to stay ahead of emerging risks.
Since its launch in 2017, DECODE has drawn an average of 1,000 attendees annually, offering a trusted platform for dialogue, training, and upskilling among cybersecurity professionals in the Philippines.
As AI’s role expands, Trend Micro stresses that consistent research, vigilant governance, and collaborative action are essential to securing the digital future. —Ed: Corrie S. Narisma
