Check Point Software Technologies has launched a new artificial intelligence security platform designed to address growing risks as AI systems evolve from passive tools into autonomous actors capable of making decisions and executing tasks within enterprise environments.
The platform, known as the AI Defence Plane, is positioned as a unified control layer that enables organisations to manage how AI systems are connected, deployed, and operated across business functions.
The launch reflects a broader shift in enterprise AI usage. While earlier deployments focused on generative tools such as chatbots and content assistants, newer systems are increasingly being designed to take action, accessing data, triggering workflows, and interacting with other systems.
“The enterprise is entering the agentic era. AI is no longer limited to generating content. It is beginning to access systems, use tools, chain actions, and operate with increasing autonomy. That changes the security model,” said David Haber, VP, AI Security, Check Point Software Technologies. “The AI Defense Plane provides that control across employees, applications, and AI agents.”
What You Need to Know
According to Haber, the key challenge is no longer limited to monitoring outputs, but controlling behaviour, ensuring that AI systems operate safely within live environments.
As organisations move AI systems from experimentation to production, the associated security risks are also expanding.
The attack surface now extends beyond prompts and models to include automated workflows, delegated actions, and non-human access points, areas that are more difficult to monitor and secure.
There is also a growing concern around “shadow AI”, where unsanctioned tools and agents operate within organisations without formal oversight, increasing the risk of data leaks and system vulnerabilities.
A Unified Security Approach
Built on Check Point’s existing AI security infrastructure, the AI Defence Plane integrates capabilities from its ThreatCloud AI platform as well as recent acquisitions, including Lakera and Cyata.
The system combines several key functions, discovery, governance, observability, runtime control, and continuous validation into a single platform designed to secure AI throughout its lifecycle.
At its core is an AI-native security engine that analyses millions of interactions and leverages real-time threat intelligence to detect and respond to risks.
The platform is designed to deliver adaptive protection in under 50 milliseconds and supports more than 100 languages, enabling organisations to respond to threats at machine speed.
Moving Beyond Model Guardrails
Unlike traditional approaches that focus on limiting what AI models can generate, Check Point’s platform is designed to secure how AI behaves in real-world environments.
This shift reflects a growing recognition that risks emerge not only from what AI systems produce, but from what they are allowed to do, particularly when integrated into business-critical workflows.
By focusing on runtime control, the AI Defence Plane aims to enforce policies where risks are most likely to materialise: within active systems and live operational processes.
“Red teaming has become essential for agentic systems. When AI can query infrastructure, trigger workflows, and interact with sensitive data, the risk is no longer theoretical. Organizations need continuous testing to understand how these systems can be manipulated, where controls break down, and how resilient they are in production,” said George Davis, Product Leader, Sierra.
As AI adoption accelerates globally, enterprises are under increasing pressure to ensure that these systems are secure, reliable, and aligned with organisational policies.
Experts say Check Point’s latest offering signals a shift towards more comprehensive AI security frameworks, ones that go beyond static protections and focus on dynamic, real-time control.
Talking Points
It is notable that Check Point is shifting the conversation around AI security from controlling outputs to controlling behaviour, which reflects the real risks emerging in enterprise environments today.
As AI systems become more autonomous, capable of accessing data, triggering workflows, and making decisions, the security challenge moves beyond what AI generates to what it is allowed to do.
At Techparley, we see this as a critical evolution. Traditional guardrails are no longer sufficient in an era where AI can act independently within live business systems.
The introduction of a unified control plane is particularly significant. Bringing discovery, governance, observability, and runtime control into one platform simplifies what is otherwise a highly fragmented and complex security landscape.
The focus on real-time protection and machine-speed response also aligns with the pace at which AI-driven threats are evolving, especially as malicious actors begin to automate attacks.
The inclusion of continuous AI red teaming is a strong addition, as it allows organisations to proactively identify vulnerabilities before they can be exploited in real-world scenarios.
——————-
Bookmark Techparley.com for the most insightful technology news from the African continent.
Follow us on Twitter @Techparleynews, on Facebook at Techparley Africa, on LinkedIn at Techparley Africa, or on Instagram at Techparleynews.
