TLDR: SPLX, a two-year-old AI security startup, has significantly enhanced its platform by introducing AI Runtime Protection, a new tool designed to safeguard AI applications and workflows in real-time against emerging threats like prompt injections and jailbreaks. The company also launched ‘Analyze with AI’ for automated red team result analysis and reported substantial Q2 2025 growth, including a 160% quarter-over-quarter increase and the addition of five new Fortune 500 clients.
SPLX, formerly known as SplxAI, a rapidly growing two-year-old startup specializing in artificial intelligence security, has announced a significant expansion of its platform with the introduction of ‘AI Runtime Protection.’ This new tool is purpose-built to secure AI application development, workflows, and real-time operations, addressing the escalating threats faced by AI systems.
AI Runtime Protection is designed to act as a real-time firewall for AI, continuously monitoring and filtering inputs and outputs to stop threats as they occur. This includes safeguarding against malicious activities such as prompt injections, jailbreaks, and social engineering attacks, which hackers leverage to compel AI models to leak sensitive data, bypass guardrails, or generate harmful content. The tool also enforces behavioral guardrails to ensure AI models comply with policies, all with near-zero latency to maintain performance and user experience.
In addition to AI Runtime Protection, SPLX has rolled out ‘Analyze with AI,’ a feature that provides security teams with comprehensive analysis and summaries of results from SPLX’s ‘Probe,’ the platform’s automated red teaming tool. This allows for a deeper understanding of vulnerabilities uncovered during continuous security testing. These new capabilities follow the release of ‘Agentic Radar’ three months prior, an open-source feature offering enhanced visibility into agentic AI workflows and their potential vulnerabilities.
Kristian Kamber, co-founder and CEO of SPLX, highlighted the evolving threat landscape, stating, “Agentic systems introduce a whole new class of vulnerabilities. In traditional software, you worry about a bad input entering at the edge. With multi-agent AI workflows, that input can propagate inside the system, from one agent to another, escalating into what we call a chained exploit.” Kamber further emphasized the shift in security paradigms, noting, “It’s a complete shift from point-in-time assessments to continuous AI security.”
The company, founded in 2023, has demonstrated impressive growth and market traction. In Q2 2025, SPLX achieved a remarkable 160% quarter-over-quarter growth and successfully onboarded five new Fortune 500 companies as customers. The firm also boasts a 100% success rate in bake-offs and champion-led Proof of Concepts (POCs) for its automated, continuous AI red teaming and security testing of AI agents. This expansion comes after SPLX secured $7 million in seed funding in March 2025, following an earlier $2 million raise in September 2024, funds earmarked for accelerating platform development and market expansion.
Also Read:
- Cycode Unveils AI Exploitability Agent to Revolutionize Vulnerability Prioritization
- Qualys Revolutionizes Cybersecurity with Agentic AI-Powered Risk Operations Center
SPLX is positioning itself as a crucial vendor in the burgeoning AI security market, especially as cybercriminals increasingly target commercial AI operations. The company will be showcasing the full capabilities of its enhanced platform at Black Hat USA 2025, hosting 1:1 meetings to demonstrate its end-to-end security solutions for the entire AI lifecycle, from development through deployment to real-time operation.
