Check Point Software Technologies has launched the Check Point AI Network Firewall, a new capability delivered through its R82.20 firewall software release that brings artificial intelligence security directly from the physical and virtual firewalls organisations already operate. The move addresses what the company calls a blind spot in enterprise networks: prompts, autonomous agent actions and sensitive business context flowing through infrastructure that traditional firewalls were never built to see.

“AI is transforming the enterprise network, and with the AI Network Firewall, we are transforming the firewall to secure it,” said Nataly Kremer, Chief Product Officer at Check Point. “The network is where every prompt, model call, and agent interaction already converges, yet traditional firewalls were never built to see or govern that activity. By bringing AI security into the firewall organisations already run, we give security teams the visibility and control to enable AI adoption safely, in real time.”

AI blind spots are already widespread

Check Point Research’s AI Security Report 2026 found that between 87% and 93% of organisations experience at least one high-risk generative AI interaction every month, while the share of prompts carrying sensitive corporate, personal or regulated data has doubled in a year to one in every 25 interactions. The research also found that organisations now run an average of ten AI applications a month, many outside any formal governance process, and that security weaknesses were present in 40% of 10,000 Model Context Protocol (MCP) servers reviewed.

Separately, Check Point Research identified 15,300 indirect prompt-injection payloads planted across public web pages, with roughly 70% hidden in parts of a page no human reader would ever see.

Turning existing firewalls into AI protection

Unlike approaches that require deploying a separate virtual firewall alongside existing infrastructure, the AI Network Firewall runs from the physical or virtual firewalls customers already operate, scaling across branches, data centres, cloud and multi-cloud environments. It covers three domains: employee AI use, including discovery of shadow and sanctioned AI tools and blocking sensitive data leakage based on prompt intent; AI Tools via MCP, covering visibility into MCP servers and enforcement across every interaction; and AI application and LLM protection, which blocks prompt injection and adversarial inputs inline before they reach the model.

The capability sits within Check Point’s AI Defense Plane, a unified control layer for discovering, governing and protecting AI activity across the network, endpoints, cloud, applications and APIs. Check Point is also extending central policy management to its SASE and SD-WAN offerings, enabling zero-trust enforcement across IT, OT and micro-segmentation tools including Illumio from a single console.

“Organizations are challenged to deploy dedicated AI security solutions, which are additive to their existing security architecture – contributing to the sprawl of dis-jointed, siloed security tools and agents,” said Pete Finalle, Research Manager, Trusted Access and Network Security at IDC. “While rare, the native integration of AI security across existing enforcement points provides improvements to visibility, telemetry effectiveness, security posture, and management simplicity.”

Chris Konrad, Vice President, Global Cyber at WWT, said policy alone will not solve shadow AI, since employees will keep using AI tools to move faster, often before security teams are aware those tools are in use. Integrating AI visibility and enforcement directly into the firewall, he said, gives teams a practical control point without slowing down innovation.

The Check Point AI Network Firewall is available now.