This announcement comes during a year of rapid innovation for Red Hat AI, following the introduction of Red Hat AI Inference Server and the release of Red Hat AI 3. Customers worldwide and across all industries are adopting Red Hat AI to drive innovation through generative, predictive, and agentic AI applications.
As companies move from experimentation to production, they face a complex challenge: deploying models that are not only powerful but also demonstrable, reliable, and secure. Security capabilities and safeguards are essential for modern machine learning operations (MLOps). This focus on security and trust reflects Red Hat and IBM's commitment to helping customers adopt a security-first mindset as they responsibly scale AI in hybrid cloud environments.
The integration of Chatterbox Labs' technology creates a unified platform where security is built in, strengthening Red Hat's ability to enable production AI workloads with any model, on any accelerator, anywhere. Addressing the unforeseen repercussions of AI: Founded in 2011, Chatterbox Labs brings critical technology and expertise in AI security and transparency. Its expertise in quantitative AI risk has been praised by global independent think tanks and policymakers, and this acquisition brings key machine learning technology to Red Hat.
Chatterbox Labs offers automated and customized AI security and protection testing capabilities, providing the objective risk metrics that business leaders need to approve AI deployments in production.
The technology offers a robust, model-agnostic approach to validating data and models through:
AIMI for Generative AI: Provides independent, quantitative risk metrics for large language models (LLMs).
AIMI for Predictive AI: Validates any AI architecture against key pillars, including robustness, fairness, and explainability.
Protection Mechanisms: Identifies and corrects unsafe, toxic, or biased prompts before models are deployed to production.
This acquisition aligns with Red Hat’s vision to support diverse hybrid cloud deployment models and environments. It also complements the future-proof capabilities introduced in Red Hat AI 3, specifically for agentic AI and the Model Context Protocol (MCP). As enterprises adopt agentic AI, reliable and secure models become even more critical, given the complex and autonomous role of AI agents and their potential impact on core enterprise systems. Chatterbox Labs has conducted research on holistic agentic security, including monitoring agent responses and detecting triggers for MCP server actions. This work aligns with Red Hat’s roadmap for Llama Stack and MCP support, positioning Red Hat to secure the next generation of intelligent and automated workloads on a trusted, enterprise-ready foundation. By combining Red Hat's MLOps capabilities with Chatterbox Labs' protection mechanisms, Red Hat will enable organizations to operationalize their AI investments with greater confidence.
