Trellix Accelerates Engineering Transformation with AI-Native Security

Cybersecurity has reached an inflection point where incremental improvement no longer holds the line against machine-speed threats. Trellix is responding by rebuilding its engineering foundations around AI-native security, embedding frontier models into every stage of the software development lifecycle.

Recent Trellix research shows a 67% increase in AI-driven APT campaigns and a 300% surge in monthly attack cadence. With frontier models like those from OpenAI and Anthropic demonstrating sandbox breakouts, the industry faces compressed time-to-exploit windows that legacy vulnerability management cannot address. Trellix is meeting this challenge head-on with an AI-native security engineering framework standardized across all engineering teams.

What was announced

Trellix CTO Joe Chen outlined how the company has transitioned from experimental AI pilots to a fully standardized AI-enabled framework covering the entire codebase, architecture, and supply chain. Frontier AI models now review the Trellix codebase on an accelerated timeline, and AI-powered auditing is embedded directly into development pipelines so vulnerabilities surface earlier in the SDLC.

To power this shift, Trellix established strategic partnerships with Anthropic and LangChain, gaining privileged access to cutting-edge models, frameworks, observability tooling, and product roadmaps. A multi-year spend commitment with Anthropic, signed in May, reflects the depth of the investment.

Why this matters

For CISOs, CIOs, IT directors, and procurement leaders, the message is direct: the organizations leading the next era of cyber defense are those willing to rebuild engineering foundations so AI is native to how they build and protect, not bolted on. Trellix's approach demonstrates how shift-left AI security can compress detection windows, reduce residual risk in the supply chain, and free engineering talent to focus on architecture and customer outcomes rather than reactive triage.

Security is not a constraint on innovation but the condition that makes it sustainable

Technical details

• AI-native engineering system: AI woven into build processes, not layered on top; features ship in smaller, highly validated increments.
• Shift-left vulnerability detection: Frontier models identify and remediate vulnerabilities earlier in the SDLC with continuous visibility.
• Secure-by-design architecture: Simplified architectural philosophy with AI-driven auditing embedded across the development pipeline.
• Anthropic partnership: Privileged access to frontier models, backed by a multi-year spend commitment.
• LangChain partnership: Frameworks and observability aligned with Trellix engineering and research teams.
• Hybrid model strategy: Both closed-source and open-source AI models used based on task suitability.

Softprom and Trellix

Softprom is the official distributor of Trellix. Enterprise customers and partners can access AI-native security solutions, XDR, EDR, and SIEM technologies through Softprom's certified team of engineers and account managers.

This content was prepared as part of the Softprom DistriFlow project — an automated system for monitoring and adapting vendor news. Original source: original article.