• June 12, 2025

Global cybersecurity leader Trend Micro Incorporated predicts a rise in highly personalized, AI-powered cyber threats by 2025 that could lead to more sophisticated scams, phishing attempts, and influence operations.

“As generative AI makes its way ever deeper into enterprises and the societies they serve, we need to be alert to the threats. Hyper-personalized attacks and agent AI subversion will require industry-wide effort to root out and address,” said Ian Felipe, Country Manager at Trend Micro.

“Business leaders should remember that there’s no such thing as standalone cyber risk today. All security risk is ultimately business risk, with the potential to impact future strategy profoundly,” he added.

One of the most alarming predictions in Trend Micro’s latest report is the rise of malicious “digital twins,” wherein a cybercriminal uses stolen personal information to train an AI model that mimics your writing style, knowledge, and even personality. When combined with deepfake video and audio, these AI-powered impersonations could be used for identity fraud or to “honeytrap” or manipulate friends, colleagues, and family members.

The report also anticipates deepfakes and AI being used to enhance existing cybercriminal tactics, including business email compromise (BEC) scams, “fake employee” schemes, and romance scams. Attackers are also expected to leverage AI for more effective open-source intelligence gathering and the creation of fake social media accounts for spreading misinformation.

Aside from AI-powered attacks, Trend Micro cautions businesses to address vulnerabilities in their own AI implementations, such as agent hijacking, unintended data leaks, and resource-draining attacks.

Traditional threats, including ransomware, API exploits, and vulnerabilities like XSS and SQL injection, also remain a concern, with ransomware tactics evolving to bypass security tools and target cloud systems and IoT devices.

Time for action

To combat these evolving cyber risks, Trend Micro recommends organizations adopt a risk-based cybersecurity approach to identify and mitigate threats proactively. Companies should utilize AI to assist with threat intelligence, attack path prediction, and remediation.

Enhancing user training and awareness to recognize AI-enabled cybercrime is also crucial, as is securing AI systems by implementing input validation, response monitoring, and layered defenses against AI abuse.

Organizations should also improve visibility into AI agents and their interactions to detect anomalies and strengthen defenses in cloud environments with Attack Path Prediction techniques.

Click here to read the full report on Trend Micro’s cybersecurity predictions for 2025.

Leave a Reply

Social