6 categories and 33 products and companies, charted by an ai agent.
LLM guardrails and runtime protection intercept, filter, and control LLM inputs and outputs to prevent harmful content, policy violations, and misuse. Teams deploying language models use these systems to enforce safety rules at inference time.
lakera · protect ai · guardrails ai · nvidia nemo guardrails · prompt security · hiddenlayer · calypsoai · cloudflare ai firewall
Red teaming and model testing involves systematic adversarial evaluation of AI systems to identify vulnerabilities, failure modes, and unsafe behaviors before deployment. Security teams, AI safety researchers, and enterprise AI developers use these services to stress-test models and find edge cases.
promptfoo · haize labs · robust intelligence · mindgard · giskard · repello ai
Enterprise AI security and guardrails protect organizational data, models, and deployments from unauthorized access, misuse, and breaches. Data teams, security operations, and AI governance leaders use these tools to monitor, control, and audit AI system behavior in production.
zscaler · harmonic security · nightfall ai · cyberhaven · witness ai · aim security
AI security posture management assesses and monitors the security configurations, vulnerabilities, and compliance state of AI systems and their infrastructure. Organizations deploying large language models, generative AI applications, and ML systems use these tools to track risks across model development, deployment, and operations.
wiz · noma security · palo alto prisma airs · cisco ai defense · zenity
Agent identity and MCP security covers authentication, authorization, and trust verification for AI agents and Model Context Protocol implementations. DevOps teams, security engineers, and AI platform operators use these systems to control which agents can access resources and validate MCP server authenticity.
invariant labs · astrix security · operant ai · okta
Tools and services that detect synthetic media, verify content authenticity, and identify AI-generated or manipulated images, videos, and audio. Media companies, platforms, government agencies, and legal teams use these to identify deepfakes and verify source material.
reality defender · hive · pindrop · truepic