Prompt Security (now part of SentinelOne)
Runtime GenAI and agentic-AI security company preventing AI-related data leakage and protecting AI agents, acquired by SentinelOne for ~$250M.
Visit Website ↗Overview
Prompt Security focused on securing generative and agentic AI at runtime — monitoring and controlling interactions between users, employees and AI tools/agents to prevent sensitive data leakage into external AI services and to protect deployed AI agents from manipulation. Its positioning is closer to a GenAI-focused extension of data-loss-prevention (DLP) than to model-layer security.
SentinelOne completed its acquisition of Prompt Security on September 5, 2025, for a total consideration of roughly $133.6M in cash plus stock and assumed options (reported deal value around $250M in some coverage), as part of SentinelOne’s strategy to extend its Singularity Platform to cover GenAI and agentic AI use cases end to end.
Innovation Matrix Assessment
Built runtime GenAI protection covering both data-leakage prevention and agent security, and was acquired within roughly two years of the generative AI security wave beginning.
Runtime interception of AI interactions gives security teams a practical DLP-style control point for GenAI usage, an immediately actionable capability.
Acquired by SentinelOne, a major public endpoint-security vendor, for a deal SentinelOne itself disclosed at roughly $133.6M cash plus equity — a strong, primary-source-confirmed validation event.
Extends the well-established DLP control pattern to GenAI traffic; valuable but conceptually adjacent to existing data-security tooling rather than an entirely new control type.
No independent, third-party efficacy validation of its detection or blocking accuracy was found.
Preventing sensitive-data leakage into external AI tools and securing agent behavior are durable needs as GenAI and agentic AI usage keeps expanding inside enterprises.
Why CISOs Should Care
Gives security and data-loss-prevention teams visibility and control over what data employees and AI agents send to external and internal GenAI tools, closing a gap traditional DLP wasn't built to cover.
What Makes It Different
Applies runtime interception and policy enforcement specifically to AI/agent traffic rather than requiring security teams to retrofit generic web/DLP proxies for GenAI use cases.
The Matrix Verdict
68/100 — INCREMENTAL INNOVATOR
Strong Performer on momentum, thanks to a SentinelOne-confirmed acquisition price, but its conceptual approach (DLP extended to AI) is scored more conservatively on disruption than category-native approaches like model-layer red-teaming.
Editorial Note: Claims vs. Verified Findings
Deal terms are corroborated by SentinelOne's own press release and SEC-related filings. Founding year and pre-acquisition headcount were not found in this research pass and are marked Undisclosed rather than estimated from general knowledge.
Sources
- SentinelOne press release — https://www.sentinelone.com/press/sentinelone-to-acquire-prompt-security-to-advance-genai-security/
- SEC Form 10-Q — https://www.sec.gov/Archives/edgar/data/1583708/000158370825000159/s-20251031.htm
- Prompt Security blog — https://prompt.security/blog/prompt-security-sentinelone-a-new-chapter-begins
Alternatives to Prompt Security (now part of SentinelOne)
Zenity
Governance and security platform for AI agents and low-code/no-code development, securing agent identity, permissions and behavior across the…
Palo Alto Networks (Prisma AIRS)
Major cybersecurity incumbent's unified AI security platform (Prisma AIRS) for discovering, assessing and protecting AI models, agents and…
Cisco AI Defense (Robust Intelligence)
AI security product line built on Robust Intelligence's algorithmic red-teaming and AI Firewall technology, now integrated into Cisco's…
Lakera
LLM security platform providing prompt-injection detection and real-time guardrails for generative AI applications, acquired by Check Point.
HiddenLayer
AI security platform unifying ML supply-chain scanning, runtime defense, posture management and automated red-teaming for enterprise AI systems.
Protect AI (now part of Palo Alto Networks)
AI/ML security company (model scanning, MLSecOps, AI-SPM) acquired by Palo Alto Networks for $634.5M and folded into Prisma…