A CYBERSECURITY company has issued a warning about the rise of highly customized, AI-powered cyber threats that could significantly amplify scams and phishing and influence operations in 2025 and beyond.
“As generative AI becomes more integrated into enterprises and society, we must remain vigilant. Hyper-personalized attacks and AI subversion will demand industry-wide collaboration to combat them. Business leaders should understand that cyber risks are no longer standalone concerns; they are business risks with far-reaching strategic implications,” said Ian Felipe, country manager at Trend Micro Philippines.
Trend Micro’s 2025 predictions report highlighted the emergence of malicious “digital twins,” where breached or leaked personal information is used to train large language models (LLMs) to mimic individuals’ knowledge, personality and writing style. Combined with deepfake video/audio and compromised biometric data, these AI-powered tools could enable identity fraud, impersonation and targeted social engineering attacks.
Deepfake technology and AI will enable large-scale, hyper-personalized cyberattacks, including enhanced business email compromise and “fake employee” scams; targeted financial scams, including “pig butchering” frauds; AI-driven romance scams that manipulate victims before passing them to human operators; improved open-source intelligence gathering for adversaries; and large-scale disinformation campaigns using fake social media personas.
Moreover, as businesses increasingly adopt AI, they must prepare for new security challenges, such as an AI agent hijacking to manipulate or exploit enterprise systems as well as unintended data leaks from generative AI tools and AI resource consumption leading to denial-of-service attacks.
Other pressing cybersecurity threats are memory corruption bugs, API exploits, container escapes and legacy vulnerabilities like XSS and SQL injection. Cybercriminals will evolve tactics to bypass endpoint detection and response tools.
Cyber defenses
To combat these escalating threats, Trend Micro recommends businesses take a proactive, risk-based approach to cybersecurity. It also suggests implementing centralized risk identification and mitigation strategies and leveraging AI for threat intelligence, asset management and attack path prediction.
The company also advised companies to enhance employee training to recognize and respond to AI-powered cyber threats and secure AI technologies against abuse through robust validation and multi-layered defenses.
It also advised the firm to strengthen LLM security via sandboxing, strict data validation and protection against prompt injection attacks; ensure supply chain security and implement multi-layered network defenses; and maintain visibility into AI-driven security risks and attack paths. / KOC