Trend Micro, a world leader in cybersecurity, announced the open source of Trend Cybertron, an AI model and an agent manager designed to accelerate the development of autonomous cybersecurity agents. As one of the first specialized LLMs in Cybersecurity, it offers organizations and researchers worldwide access to advanced cybersecurity capacities. The specialized trendy cybertron model is refined using LLAMA 3.1 and supports a rapid and reliable deployment with NVIDIA NIM inference microservices on an accelerated NVIDIA infrastructure.
World organizations have trouble innovating and developing while being increased by security challenges, fragmented ad hoc solutions and a crushing flow of threat alerts. Complexity requires a passage to a proactive approach. Propelled by Nvidia AI at the heart, Trend Cybertron goes beyond the continuation of threats, applying intelligent and decision-making agents to predict and react.
To operationalize this vision, Trend has built its agentic AI strategy with NVIDIA AI software to accelerate the automation of cybersecurity, helping organizations to manage the risks with the digitization of resources, risk assessment, priorities based on priorities and sanitation suggestions.
Organizations can apply several plans and transparently integrate AI agents to automate security tasks with resource digitization, proactively mitigating threats and setting their defenses to manage threats more effectively, all in the Nvidia ecosystem. More specifically, this will help you:
- Improve the security posture: allowing teams to precisely anticipate the risks on the entire surface of the attack.
- Reduce alert overload: relieve fatigue from Secops teams thanks to a more precise hierarchy.
- Save developers time: overcome the shortages of safety skills by providing usable information to help them identify and correct risks.
- Free up greater value: provide more powerful information from existing risk sensors.
Trend Cybertron is designed to proactively manage risks, taking advantage of the intelligence of threats of more than 250 million sensors in the world – the widest in industry. He interprets user requests, generates usable plans and performs a holistic risk assessment by recovering cybersecurity intelligence in real time from the Trend Micro cloud, ultimately providing tailor -made recommendations and best practices to secure the AI systems of a company.
Nvidia has provided support and AI microservices to develop and deploy the model. Micro-formed trend and optimized the cybertron model trend for inference using the NVIDIA DGX supercaluation in order to reduce the time necessary to refine the model.
Trend Cybertron is currently made up of an ia model of $ 8 billion and an initial specialized AI agent, with developing models and agents to extend its cybersecurity capacities. A larger and more advanced version with 70 billion model parameter is planned for a near future to meet the challenges of the cybersecurity of tomorrow.
Kevin Simzer, COO at Trend Micro
The Cybertron trend secret sauce is the data that she learns permanently, refined for detection and attenuation of optimized threats. By providing the highest quality threat data and the expertise of the head of the NVIDIA industry, we have made proactive security a reality, allowing us to predict and prevent threats like never before. This innovation is not only a victory for our customers – it is a question of making the whole digital world and connected a safer place.
Pat Lee, vice-president of business strategic partnerships at NVIDIA
With the ability to understand, reason and act, AI agents offer organizations a new powerful cybersecurity tool. AI agental security agents built with the model and the Trend Cybertron frame using Nvidia AI can analyze massive amounts of data in real time to detect potential threats, adapt dynamically and respond independently.