Red Hat has been hard at work expanding its AI portfolio this year. A few months ago, the company launched Red Hat AI 3, which supports agent AI and Model Context Protocol (MCP). And before that, we added Red Hat AI Inference Server, RHEL AI, and Red Hat AI third-party validated models to our product list.
Now, it has announced the acquisition of London-based AI safety company Chatterbox Labs. Red Hat says We plan to open source our technology in the futurefollows the usual approach of acquiring proprietary technology and releasing it to the community.
Red Hat's latest purchases
Founded in 2011, Chatterbox Labs has been testing AI risk and safety for more than a decade. built by the company AIMI platformprovides model-independent validation for AI systems.
This platform has three main services. AIMI for Gen AI provides quantitative risk metrics for large language models (LLMs). AIMI for Predictive AI validates the robustness, fairness, and explainability of traditional AI architectures. and guardrail Catch unsafe, harmful, or biased prompts before the model goes into production.
Additionally, the platform can proactively probe models for vulnerabilities such as prompt injection, jailbreak, and data leakage.
For existing Red Hat customersThis acquisition means safety tools will be built into the platform rather than being bolted on afterwards. This should enhance Red Hat's existing AI products and add “.AI securityStamp your portfolio.
Steven Huels, vice president of AI engineering and product strategy at Red Hat, said of the acquisition:
Enterprises are rapidly moving AI from the lab to production, increasing the urgency for reliable, secure, and transparent AI deployments. Chatterbox Labs' innovative, model-agnostic safety testing and guardrail technology is the critical “security of AI” layer the industry needs.
By integrating Chatterbox Labs into the Red Hat AI portfolio, we strengthen our promise to our customers to provide a comprehensive open source platform that not only allows them to run any model anywhere, but also with the confidence that safety is built in from the beginning. This acquisition will enable truly responsible, production-grade AI at scale.
Recommended reading 📖: Linux Foundation launches Agentic AI Foundation
Linux Foundation launches Agentic AI Foundation with industry-wide support
Anthropic's MCP, Block's goose, and OpenAI's AGENTS.md form the foundation of this new effort.
