A pioneer in AI safety, Chatterbox provides a robust, model-agnostic approach to validating data and large language models.

Chatterbox Labs delivers automated, customised AI security and safety testing capabilities, providing the factual risk metrics that enterprise leaders need to approve the deployment of AI to production.

The technology offers a robust, model-agnostic approach to validating data and models through:

AIMI for gen AI: Delivering independent quantitative risk metrics for Large Language Models (LLMs).

AIMI for predictive AI: Validating any AI architecture across key pillars, including robustness, fairness and explainability.

Guardrails: Pinpointing and remedying insecure, toxic, or biased prompts before putting models into production.

Chatterbox was created in Queen Mary’s School of Electronic Engineering and Computer Science, and has now successfully exited to Red Hat, the world’s leading provider of open-source solutions.

By joining Red Hat, Chatterbox can bring its validated, independent safety metrics to the open-source community, so we can all benefit from AI that is secure, scalable and open.