Organisations deploying AI workloads are creating and retaining larger volumes of sensitive data across industrial, medical, transportation and edge computing environments. As AI applications expand beyond proof-of-concept projects into operational systems, protecting stored data has become a growing challenge for embedded hardware designers.
According to Apacer, increasing use of AI models, machine learning datasets and edge inference platforms is creating new risks around data exposure, unauthorised access and intellectual property theft. These concerns are extending beyond cloud infrastructure into the storage devices embedded within equipment operating at the network edge.
The issue is becoming more significant as organisations face a broader cyber threat landscape. The Department for Science, Innovation & Technology recently highlighted that cyberattacks are estimated to cost the UK economy £14.7 billion annually, while 43% of businesses reported security breaches during the previous year, citing figures from the Department for Science, Innovation & Technology. Hardware-level security is increasingly being considered alongside traditional software-based protection measures.
Apacer has developed CoreSecurity2 to address security risks associated with data storage in AI-enabled systems. The technology is designed to protect sensitive information stored on industrial SSDs and memory products, helping prevent unauthorised access to critical data throughout the device lifecycle.
The approach reflects a wider movement towards security mechanisms implemented directly within hardware, with an industry shift towards security designed into silicon architectures from the outset rather than relying solely on software controls.
Apacer’s CoreSecurity2 technology incorporates the following features:
- Destroying – irreversibly disables SSD usage and eliminates all data when devices are compromised or decommissioned.
- Erasing – delivers complete data sanitisation with quick, full, and military-grade erase modes: ensuring zero data remanence.
- Encrypting – implements AES-256 encryption with TCG Opal 2.0 compliance for secure access control and instant data invalidation.
- Protecting – prevents unauthorised data modification via write-protect and device-lock mechanisms: no additional software required.
These capabilities are intended to help protect sensitive AI training data, model parameters, operational records and proprietary algorithms stored within embedded systems.
The increasing value of AI-generated and AI-dependent data is driving demand for storage solutions capable of supporting both performance and security requirements. For industrial equipment manufacturers, medical device developers and transportation system designers, storage security is becoming a design consideration earlier in the product development cycle.
As an authorised distribution partner for Apacer, Astute provides access to the manufacturer’s industrial memory and storage portfolio alongside engineering support and supply chain services.
The company has continued to expand its portfolio of technologies focused on cyber resilience and hardware-rooted security. Recent partnerships and product initiatives have focused on reducing exposure to cyber threats through secure-by-design approaches, including memory-safe processing architectures and advanced data protection technologies.
As AI adoption accelerates across embedded and industrial markets, storage security is becoming a more prominent requirement alongside performance, endurance and long-term availability. Technologies such as CoreSecurity2 illustrate how storage manufacturers are responding to growing concerns around protecting valuable data assets at the hardware level before they become targets for compromise.
Damian Semple, Franchise Marketing Manager, comments: “As AI deployments grow, organisations are storing more valuable intellectual property and operational data at the edge. Security requirements are increasingly influencing storage selection decisions alongside endurance, lifecycle support and long-term availability, making early engagement with trusted technology partners more important during the design phase.”
