AI security
For companies building chatbots and copilots or rolling out AI agents to staff.
- AI governance programAI inventory, acceptable-use policy and risk register aligned to NIST AI RMF and ISO/IEC 42001, with Texas TRAIGA in mind.
- LLM application assessmentTesting against the OWASP Top 10 for LLM applications: prompt injection, data leakage, insecure output handling.
- AI agent and MCP security reviewTool permissions, identity, secrets and audit logging for agents, MCP servers and gateways.
- AI red teamingAdversarial testing of a model or agent before launch, with a prioritized findings report.
- Shadow AI reviewWhich AI tools your staff use, what data reaches them, and which vendors pass review.