Check Point Software Technologies has rolled out its Agentic Network Security Orchestration platform, a purpose‑built autonomous‑agent architecture that aims to shoulder the bulk of network security operations for enterprises without the need for round‑the‑clock human oversight. The launch comes as South African organisations grapple with ever‑expanding hybrid clouds, a surge in connected devices and the relentless drift of policies that outpace the capacity of traditional security teams.
Enterprises across the country are feeling the strain. A single change request can now take two to four weeks to navigate analysis, security review and policy dependencies, only to break another service and trigger a fresh cycle of fixes. Segmentation projects linger on board agendas for years, and policy drift becomes inevitable when workloads move faster than any security team can keep up. The result is a stalled zero‑trust journey and exposed attack surfaces that could be exploited by increasingly sophisticated threat actors.
Check Point’s new platform tackles these challenges on three fronts that have traditionally held back network security management. First, it replaces thousands of static firewall rules with intent‑based policies that translate business language into enforceable controls. Second, it swaps fixed threat‑prevention profiles for dynamic, exposure‑based controls that adapt in real time. Third, it consolidates fragmented vendor consoles into a single orchestration layer that spans the entire network, allowing security teams to set high‑level business intent while autonomous agents handle the granular execution.
How Agentic Network Security Orchestration Redefines Autonomous Execution
“For the first time, security teams can operate entirely at the level of business intent,” explained Jonathan Zanger, Chief Technology Officer at Check Point. “With Agentic Network Security Orchestration, teams define what needs to be protected and what the policy should achieve. Everything below that – rule creation, policy tightening, virtual patching – is handed to AI agents to execute autonomously, within predefined guardrails and under continuous human oversight.”
At the heart of the platform lies a proprietary Network Knowledge Graph, a live relational model that mirrors a customer’s actual environment – topology, traffic flows, asset dependencies and configuration data are refreshed continuously. Unlike generic AI tools that rely on static training sets, Check Point’s agents reason over this live graph, grounding every decision in the organisation’s real‑time reality.
The semantic intelligence layer goes a step further, interpreting the business intent behind legacy firewall policies that may have been written years ago. Once that intent is understood, the agents act across four core capabilities:
| Core Capability | What It Does | Benefit to Enterprises |
|---|---|---|
| Intent‑to‑Policy Translation | Converts natural‑language business requirements into hardened, risk‑validated firewall rules across multi‑vendor environments. | Cuts rule‑creation time from weeks to minutes, ensuring policies match strategic goals. |
| Zero‑Trust & Policy Tightening | Continuously analyses live traffic to spot shadow access and over‑permissive settings, applying validated tightening without breaking connectivity. | Accelerates zero‑trust adoption and reduces attack surface exposure. |
| Autonomous Troubleshooting | Performs multi‑step reasoning across topology, policy history and logs to diagnose failures autonomously. | Lowers mean time to resolution from hours to minutes. |
| Continuous Compliance | Maps every rule and change to frameworks such as PCI‑DSS, NIST and DORA in real time, automating enforcement. | Replaces annual audit fire drills with continuous, auditable compliance. |
The table makes clear that each capability is designed to shave time, reduce risk and free security personnel to focus on strategic work rather than fire‑fighting.
Security teams retain authority at the intent level, approving high‑impact changes before they are executed and monitoring every agent action through a full execution trace. Check Point draws on more than 30 years of operational expertise protecting over 100 000 organisations, fine‑tuning agent skills to handle edge cases and complex configurations that generic models simply cannot manage.
Accelerating the Road Map with Deepchecks Acquisition
In a move that underlines its commitment to rapid innovation, Check Point announced the acquisition of Deepchecks, a production‑grade platform that unifies evaluation, observability, testing and monitoring for AI‑driven security agents. Deepchecks’ team, graduates of Israel’s elite Talpiot programme, brings cutting‑edge LLM expertise that will speed up the rollout of additional agents, skills and multi‑vendor support slated for the second half of 2026.
“Any multi‑agent system must include a robust evaluation layer that enables continuous measurement, tuning and improvement over time,” said Ofir Korzenyak, Vice President of AI Technologies at Check Point. “Deepchecks’ talent and intellectual property will significantly accelerate the execution of the Agentic Network Security Orchestration road map.”
Immediate Availability and Early Access
Check Point’s Agentic Security Management suite is already live. Core modules – Policy Auditor (prevents policy drift), Policy Insights (drives zero‑trust tightening) and AI Assist (speeds up daily admin tasks) – are available immediately. The Playblocks Agents component is being rolled out through an Early Availability programme, with a broader customer preview expected later this year.
South African businesses looking to modernise their network security posture can request access via the official Check Point website. The platform promises not only faster project delivery but also an auditable, continuously improving security fabric that aligns with stringent local regulations and global compliance standards.
As the digital landscape in South Africa grows more complex, the shift from static, manually‑managed firewalls to an Agentic Network Security Orchestration model could prove decisive. By letting AI agents handle the heavy lifting while security leaders steer with clear business intent, organisations stand to close the gap between ambition and reality, delivering robust, future‑proof protection at scale.