SplxAI

SplxAI, which emerged from stealth in March 2025 with a $7 million seed round led by LAUNCHub Ventures, builds automated red-teaming tools that defend AI agents against prompt injection and related attacks. Its platform runs simulated adversarial attacks against an organization’s AI agents and continuously monitors for new threats through log analysis, aiming to find weaknesses before attackers do rather than after an incident.

The company has disclosed named customers including KPMG, Infobip, Brand Engagement Network, and Glean, and reports 127% quarter-over-quarter growth since its platform launched in August 2024 — a company-reported figure that should be read as directional rather than audited. SplxAI also released Agentic Radar, an open-source tool that maps dependencies in agentic AI workflows and flags security gaps through static code analysis, which gives the broader security community visibility into its methodology beyond its commercial product.

What differentiates SplxAI is its focus specifically on agentic AI systems — where attack surfaces include tool calls and multi-step reasoning chains, not just a single prompt — rather than generic LLM guardrails. As a seed-stage company less than two years old, its momentum and efficacy evidence is still largely limited to self-reported growth and a small set of named enterprise logos.