We are very pleased to partner with a Modern, Cloud Native, Zero Trust CNAPP innovator like AccuKnox. Zero Trust security is a Clint Health imperative ....
Spike in AI attack vectors demands for AI Security
ModelArmor helps finding Risks with NVIDIA microservices, Exposed models, configuration, GPU/CUDA compromise. Intelligence to handle threat model and Observability in file, process, network events.
*Opensource powered by KubeArmor
One of the first AI Opensource engine to supress AI attacks at runtime
- ModelArmor leverages KubeArmor as a sandboxing engine to securely constrain the execution of untrusted models within predefined boundaries.
- AI/ML models run as processes, allowing untrusted models in AI environments poses significant risks, such as cryptomining attacks exploiting GPUs or remote command injection vulnerabilities.
- The gathered telemetry allows identifying issues such as Cyprotojacking Prevent the exploitation by enforcing pre-emptive policies and hence offers an effective framework to safeguard and enforce constraints on the models’ execution environment.
What Problems does ModelKnox solve?
ModelArmor helps in minimizing several critical problems in executing untrusted AI/ML models can be addressed as well as reduces risks associated with running untrusted models but also strengthens the overall security posture of AI environments.
Mitigation of Cryptomining Attacks
Prevents unauthorized use of GPUs for cryptomining by restricting model processes to only execute authorized computations.
Defense Against Remote Command Injection
Blocks untrusted models from executing malicious commands that could compromise the system or network.
Containment of Untrusted Code
Ensures untrusted models are isolated within a secure environment, preventing them from accessing unauthorized resources or data.
Prevention of Resource Abuse
Controls resource usage (CPU, memory, disk) to prevent models from overwhelming the system and affecting other workloads.
Reduced Risk of Data Breaches
Constrains the model’s execution to prevent unauthorized access or exfiltration of sensitive data.
Proactive Threat Mitigation
Detects and stops potential security violations before they can cause harm, via KubeArmor’s preemptive policies.
Regulatory Compliance
Helps meet security and compliance requirements by ensuring models are executed within strict operational boundaries.
Securing CUDA Libraries
Whitelist process or binaries to access CUDA libraries and denying access to all other malicious or unauthorized access attempts to exploit CUDA libraries.
Importance of model security in today’s AI-driven landscape
In today’s AI driven world, as the mundane becomes automated we must also think about the data being used by the models to implement the automation. AI systems have access to sensitive data information for various tasks. This could be valuable intellectual property, PII (Personal Identity Iformation) or insider assets that causes a huge loss when leaked to the public.
Without proper security, they could be misused for ex-filtration of the data or holding organization for ransomware in the worst case. Protecting models ensures they work safely, reliably, and follow rules, keeping systems and data secure. Hence, AI model security is an essential part of an AI-driven landscape to safeguard against these issues.
Origin Story and System Design
As AI started booming, getting placed in every vertical to automate tasks, AccuKnox also implemented an AI chatbot to help with security. Soon, we understood from our clients that there were many verticals in the industry where LLMs capability was being utilized for respective use-cases. They were essentially worried about the security of data as the AI ingested and analyzed them for insights.
ModelArmor sets out to create a solution that could protect the AI models themselves and alleviate the security concerns regarding the usage of AI. Thus ModelArmor was created to provide security focused on AI models by making use of the powerful detection and prevention capabilities of KubeArmor under the hood.
Unique Differentiator
- ModelArmor is easy to deploy with negligible process overhead as it leverages the inbuilt capabilities in the linux kernel such as eBPF and LSMs for enforcing the security policies.
- Provides an easy deployment model and scales together with the nodes on a distributed setup reducing maintenance complexity.
- Ability to define and enforce granular policies for inline mitigation sets ModelKnox apart from other alternatives which provide only detection capabilities.
ModelArmor Demonstration
Trusted By Global Innovators
- Schedule 1:1 Demo
- Product Tour
On an average Zero Day Attacks cost $3.9M
4+
Marketplace Listings
7+
Regions
33+
Compliance Coverage
37+
Integrations Support
