12 companies tracked in this category. Names and one-line descriptions are free to browse — pricing model, funding, founding year, headquarters, and cited sources are in the full dataset.
| Company | Description |
|---|---|
| Adversa AI | Adversa AI offers a continuous, automated AI red-teaming and threat-modeling platform that tests LLMs, AI agents, and GenAI applications for vulnerabilities like jailbreaks, prompt injection, and goal hijacking. |
| CalypsoAI | CalypsoAI provides a full-lifecycle enterprise platform that red-teams and defends AI models and applications at the inference layer, protecting against prompt injection, jailbreaks, and sensitive data leakage. |
| Cranium AI | Cranium AI is an enterprise AI security and governance platform that discovers and inventories AI systems, red-teams and tests models for vulnerabilities, and manages AI risk, compliance, and third-party AI exposure. |
| Giskard | Giskard provides an open-source library and enterprise platform for continuous red-teaming, vulnerability scanning, and quality/security evaluation of LLMs and AI agents, including hallucination detection and bias evaluation. |
| Haize Labs | Haize Labs builds automated red-teaming and stress-testing tools that probe large language models and AI agents at scale to surface safety, security, and reliability failure modes before deployment. |
| HiddenLayer | HiddenLayer offers an AI/ML security platform covering asset discovery, model and supply-chain scanning, automated adversarial red-teaming, and runtime protection against prompt injection and other AI-specific threats. |
| Lakera | Lakera provides an AI-native security platform (Lakera Guard, built on its Gandalf engine) delivering real-time prompt-injection defense, jailbreak detection, data-leakage protection, and red-teaming for enterprise generative AI applications. |
| Mindgard | Mindgard is an automated AI red-teaming and dynamic application security testing (DAST-AI) platform that discovers attack surfaces, runs continuous adversarial attacks, and provides runtime protection for AI systems and agents. |
| Prompt Security | Prompt Security provides a GenAI security platform delivering real-time runtime protection against prompt injection, jailbreaks, and AI-related data leakage across enterprise LLM and AI agent usage. |
| Protect AI | Protect AI built an end-to-end AI/ML security platform (Guardian, Recon, Layer) for scanning models and the AI supply chain, red-teaming AI applications, and runtime protection against AI-specific threats. |
| Repello AI | Repello AI provides an automated AI red-teaming engine (ARTEMIS) plus runtime guardrails and MCP monitoring that test and protect LLMs, AI agents, and MCP integrations against prompt injection, jailbreaks, and data exfiltration. |
| Robust Intelligence | Robust Intelligence built an AI application security platform that automated validation and runtime protection to detect and prevent security and safety vulnerabilities in AI/ML models. |