Promptfoo Raises $18.4M Series A to Build the Definitive AI Security Stack
We raised $18.4M from Insight Partners with participation from Andreessen Horowitz. Funding will accelerate development of the most widely adopted AI security testing solution..

Latest Posts

Evaluating political bias in LLMs
How right-leaning is Grok? We've released a new testing methodology alongside a dataset of 2,500 political questions..

Join Promptfoo at Hacker Summer Camp 2025
Promptfoo will be attending the AI Summit, Black Hat, and DEF CON.

AI Red Teaming for complete first-timers
A comprehensive guide to AI red teaming for beginners, covering the basics, culture building, and operational feedback loops.

System Cards Go Hard
Understanding LLM system cards and their importance for responsible AI deployment.

The Promptfoo MCP Proxy: Enterprise MCP Security
Learn about the security risks introduced by MCP servers and how to mitigate them using the Promptfoo MCP Proxy, an enterprise solution for MCP security..

OWASP Top 10 LLM Security Risks (2025) – 5-Minute TLDR
Learn the 10 biggest LLM security risks and practical fixes, in a 5-minute TLDR.

Promptfoo Achieves SOC 2 Type II and ISO 27001 Certification: Strengthening Trust in AI Security
Promptfoo achieves SOC 2 Type II and ISO 27001 compliance, demonstrating enterprise-grade security for AI red teaming and LLM evaluation tools..

ModelAudit vs ModelScan: Comparing ML Model Security Scanners
Compare ModelAudit and ModelScan for ML model security scanning.

Harder, Better, Prompter, Stronger: AI system prompt hardening
Learn essential techniques for hardening AI system prompts against injection attacks, unauthorized access, and security vulnerabilities.