Trend Micro announced its collaboration with Nvidia to develop new cybersecurity tools leveraging artificial intelligence that are designed to safeguard data centers where AI operations are conducted. These tools, showcased at the Computex conference in Taiwan, will run on Nvidia’s chips, detecting intruders and ensuring data visibility is restricted to authorized users only.

As businesses increasingly train AI systems to assist employees with tasks like answering HR questions or aiding customer service agents, they often consolidate data from various sources into a single location, making it an attractive target for hackers. “Hackers infiltrate the enterprise and find this massive honeypot of information,” Trend Micro Chief Operating Officer Kevin Simzer explained.

In addition to intrusion detection, Trend Micro aims to prevent hackers from accessing sensitive data fed into AI systems. Modern chatbots often handle user queries containing confidential corporate secrets or private customer data.

“They’re often narrowing the scope of a chatbot’s responses by providing very specific information,” Simzer noted. “Our goal is to identify and secure this information, ensuring it remains accessible only to authorized individuals and applications.”