November 20, 2026

Check Point launches AI platform to automate enterprise network security management

AI-Assist-2

Cybersecurity firm Check Point Software has launched a new AI-powered platform designed to automate the day-to-day management of enterprise network security, as it bets that human teams can no longer keep pace with the complexity of modern infrastructure.

The Agentic Network Security Orchestration Platform uses autonomous AI agents to handle tasks that have traditionally consumed security teams’ time: translating security policies into firewall rules, identifying and closing configuration gaps, troubleshooting network failures, and managing compliance across multi-vendor environments.

The launch comes as enterprise IT teams face mounting pressure from the growing complexity of hybrid cloud environments, remote work infrastructure, connected devices, and AI systems operating across their networks. Check Point argues that this complexity has pushed manual security management past its practical limits, with change requests taking weeks, Zero Trust deployments stalling, and security policies drifting as networks evolve faster than teams can track.

How the platform works

The system is built around a proprietary Network Knowledge Graph, a live model of each customer’s actual network environment that is continuously updated with topology data, traffic flows, asset dependencies, and configuration changes. AI agents use this graph to reason about the specific customer environment rather than relying solely on pre-trained models.

The platform’s four main functions cover intent-to-policy conversion (turning plain-language business requirements into validated firewall rules), autonomous policy tightening (identifying over-permissive access configurations and applying fixes), network troubleshooting, and continuous compliance monitoring against frameworks including DORA, PCI-DSS, and NIST.

Security administrators retain oversight and approval rights over significant changes, with full audit trails of agent actions.

Charlotte Wilson, head of enterprise for the UK & Ireland at Check Point, said the platform marks a shift in how security operations are structured: “Teams define what needs to be protected and what the policy should achieve. Everything below that is handed to AI agents to execute autonomously, within predefined guardrails and under continuous human oversight.”

Acquisition adds LLM expertise

Alongside the platform launch, Check Point has agreed to acquire Deepchecks, an Israeli AI company specialising in LLM evaluation and monitoring. The Deepchecks team, which includes graduates of Israel’s Talpiot advanced technology programme, will join Check Point to accelerate development of the agentic platform and build the evaluation infrastructure needed to validate agent behaviour in live enterprise environments.

The acquisition addresses a practical concern with autonomous AI systems in security contexts: how organisations can verify that agents are behaving as intended and identify when they are not.

Availability

A number of the platform’s capabilities are already available to customers, including Policy Auditor for detecting policy drift, Policy Insights for Zero Trust configuration improvements, and AI Assist for routine administrative tasks. The broader Playblocks Agents system is currently in early access, with a wider rollout including additional agents and expanded multi-vendor support planned for the second half of 2026.

Check Point, which is listed on NASDAQ and counts more than 100,000 organisations among its customers globally, said the platform’s underlying AI models have been trained on more than 30 years of operational data from its customer base.

The company faces competition in the agentic security space from a range of established vendors and startups, though it argues its combination of a live network knowledge graph, proprietary training data, and integrated evaluation layer differentiates the platform from generic AI applied to security problems.