AI Governance & Security

Credo AI is a comprehensive enterprise AI governance platform designed to help organizations discover, assess, and monitor AI agents, models, and applications. It serves Global 2000 enterprises by providing a centralized 'control plane' that bridges the gap between technical AI development and regulatory compliance, distinguishing itself through a continuous, context-aware governance loop rather than static audits.

Arthur AI is a comprehensive AI performance and governance platform designed for enterprise teams to monitor, evaluate, and secure machine learning models and generative AI agents. It distinguishes itself through a federated 'Data Plane' architecture that allows organizations to run complex evaluations and guardrails locally within their own VPC, ensuring sensitive data never leaves their secure environment.

Fiddler AI is an enterprise-grade AI Observability and Security platform designed for data scientists and governance teams to monitor, explain, and secure predictive ML models and generative AI agents. Its key differentiator is its 'AI Control Plane' which provides unified visibility across the entire agentic lifecycle—from thought to execution—combined with real-time guardrails and auditable governance.

CalypsoAI provides an enterprise-grade AI security and governance platform designed to protect Large Language Model (LLM) interactions through a real-time 'Inference Perimeter.' It empowers security and AI teams to prevent data leakage, block prompt injections, and ensure regulatory compliance across any model or vendor. Its key differentiator is its proactive, agentic red-teaming combined with real-time adaptive guardrails that secure AI applications without degrading performance.

HiddenLayer is an enterprise-grade AI security platform designed to protect machine learning models, datasets, and autonomous agents from adversarial attacks and supply chain risks. It serves CISOs and AI engineers by providing a non-invasive security layer that detects threats like prompt injection and model theft without requiring access to raw training data or proprietary model weights.

Robust Intelligence is an end-to-end AI security and governance platform designed to protect enterprises from operational, ethical, and security risks throughout the model lifecycle. Recently acquired by Cisco, it differentiates itself by offering the industry's first 'AI Firewall' and automated 'Algorithmic Red Teaming' to proactively identify and mitigate vulnerabilities like prompt injection and data drift.

Galileo is an enterprise-grade Generative AI evaluation and observability platform designed for AI engineers and data scientists to build reliable LLM applications. It differentiates itself through proprietary 'Guardrail Metrics' and a unified workflow that spans from prompt engineering and experimentation to real-time production monitoring and automated protection.

Patronus AI is an automated evaluation and security platform designed for enterprise LLM applications, helping developers detect hallucinations and safety risks at scale. It differentiates itself through proprietary, research-backed evaluation models like Lynx and FinanceBench that outperform GPT-4 in specialized scoring tasks.

Guardrails AI is an open-source framework and enterprise platform designed to ensure the reliability, safety, and compliance of LLM applications through programmable validation. It allows developers to enforce structured outputs and policy-based constraints (guards) on AI interactions, serving as a critical reliability layer for engineers building production-ready GenAI.

Lakera is an AI-native security platform designed to protect enterprise GenAI applications and agents from real-time threats like prompt injection and data leakage. It serves developers and security teams by providing a low-latency security layer that sits between users and LLMs, distinguished by its 'Gandalf' dataset which leverages insights from over 1 million hackers to stay ahead of emerging exploits.

Promptfoo is an open-source, local-first framework designed for testing, evaluating, and red-teaming LLM prompts and applications. It enables AI engineers to move beyond subjective 'vibe checks' by providing a systematic, code-centric approach to ensuring model outputs are safe, accurate, and performant across different providers.

Protect AI is a comprehensive MLSecOps platform designed to secure the entire AI lifecycle, from model development and supply chain to runtime deployment. It serves enterprise security and data science teams by providing automated vulnerability scanning, red teaming, and real-time threat detection, uniquely powered by the world’s largest community of AI security researchers.