CheckPoint Software (NASDAQ: CHKP) announced on September 16, 2025, the acquisition of AI security specialist Lakera in a strategic move that positions the cybersecurity giant to deliver comprehensive protection for enterprise AI deployments. This acquisition establishes CheckPoint's entry into the rapidly growing AI security market by bringing together Lakera's specialized inference protection technology with CheckPoint's proven Infinity Platform architecture.
The timing reflects the urgent need for AI-specific enterprise security as generative AI adoption accelerates. As CheckPoint CEO Nadav Zafrir explained, "AI is transforming every business process, but it also introduces new attack surfaces. We chose Lakera because it brings AI-native security, superior precision, and speed at scale."
Lakera's AI-Native Security Foundation
Founded in 2021 by former Google and Meta AI experts, Lakera distinguished itself by building security solutions specifically designed for AI environments rather than adapting traditional cybersecurity tools. The ~70-person company operates from Zurich and San Francisco, focusing on protecting large language models and AI agents at the critical inference layer where models process inputs and generate outputs.
Lakera's standout achievement is Gandalf, an AI security testing platform with over 80 million attack data points crowdsourced from security researchers globally. The platform serves both as a research tool for identifying AI vulnerabilities and a commercial service for Fortune 500 clients. The company's two core products, Lakera Red and Lakera Guard, provide pre-deployment security assessments and real-time runtime protection respectively.
Deal Structure and Strategic Rationale
Deal terms were not disclosed, but likely reflected a mix of cash and stock and a low hundred million dollar price tag, based on recent AI cybersecurity deals. The acquisition is set to close in Q4 2025. Lakera's Zurich headquarters will become CheckPoint's global AI security R&D center, with the leadership team remaining to lead the integration effort.
CheckPoint's acquisition thesis centers on traditional cybersecurity tools being fundamentally inadequate for AI workloads. AI systems create entirely new attack surfaces through their ability to process natural language and generate dynamic outputs, enabling novel threats like prompt injection attacks that can manipulate models to reveal training data or bypass security controls.
The strategic value lies in Lakera's AI-native approach. Unlike retrofitted solutions, Lakera's platform understands AI model behavior, recognizes adversarial inputs, and enforces policy guardrails without degrading performance. CheckPoint maintains Lakera's model-agnostic architecture, ensuring enterprises can secure any AI deployment regardless of technology stack.
Bottom Line
CheckPoint's Lakera acquisition reflects the red-hot cybersecurity M&A market in 2025, where inference-layer AI protection has emerged as the must-have capability for platform players. This acquisition spree makes strategic sense. Building AI-native security capabilities internally would take years, while startups like Lakera have already solved the hard problems around prompt injection detection, model manipulation prevention, and adversarial testing at scale. The likely low hundred million dollar price tags reflect the premium buyers are willing to pay for proven technology and immediate time-to-market advantages.
We'll be watching closely as these acquisitions play out. The real test comes in product integration and customer adoption over the coming months. CheckPoint's track record suggests they'll execute well, but with every major platform now racing to build comprehensive AI security stacks, the competitive pressure will be intense. Expect more AI cybersecurity acquisitions as the market heats up and the race for AI security dominance accelerates.