Adversa AI was founded in 2021 by Alex Polyakov and Eugene Neelou and is based in Israel, working on applied security for AI systems rather than traditional application security. Gartner has recognized the company in its AI security research, and Adversa has won industry awards including a Stevie Award and a Fortress Cybersecurity Award for its red-teaming platform.
The platform runs continuous, automated red-teaming against generative AI applications, autonomous agents, and MCP-based agent stacks, simulating the kinds of prompt injection, jailbreak, and data-exfiltration attacks a human red team would otherwise need to run manually. Because it targets agentic AI stacks specifically rather than only chatbot-style LLM interfaces, it covers a newer attack surface most legacy application-security tools weren't built to test.
Adversa targets enterprise security teams, AI governance functions, and organizations deploying LLM-based agents into production that need ongoing validation rather than a one-time penetration test. The company doesn't publish self-serve pricing; engagements are scoped through direct contact with the company.
