Machine learning (ML) has emerged as a transformative technology across various industries, including the U.S. Navy. Recently, in an effort to address a crucial privacy concern related to the security of real-time video feeds, the U.S. Navy embarked on a groundbreaking project that sought to deploy a facial recognition system that was capable of identifying VIPs without looking at raw exposed video feed.
To tackle this complex project, Protopia AI deployed Stained Glass Transform™, which played a pivotal role in the successful execution of the U.S. Navy Trident Warrior exercise.
A new category of patented privacy-enhancing technology, Stained Glass Transform™ enables AI/ML algorithms to operate accurately using holistically transformed and randomized representations. Over the past two years, Protopia AI has tested the product across a variety of use-cases and data types with leading financial services and global technology providers.
This article answers questions about how Protopia AI created a practical solution for the U.S. Navy that provides both accuracy and protection, and discusses how privacy-preserving AI and machine learning can be beneficial for a wide range of other industries as well.
Protopia AI with the US Navy.
Protopia AI partnered with NetApp and implemented Stained Glass Transform™ to demonstrate the breakthrough capability of being able to perform a task like real-time face recognition of VIPs without exposing unprotected images. A majority of solutions are not easy to incorporate as they are hardware intensive, inconsistent during the ML lifecycle, and often had prolonged delay.
The solution transformed subjects’ faces into Randomized Re‑Representations that were only readable by the facial recognition model with near‑perfect accuracy to identify VIPs. The use case showcased Stained Glass’ ability to protect real-time AI deployment successfully. Stained Glass is trained on the requirements of the facial recognition model and devises a one-way Randomized Re-Representation that is only be understood by the AI model.
The outcome of implementing the Stained Glass Transform™ during the U.S. Trident Warrior exercise was successful, as the technology enabled the identification of VIPs with precision and efficiency. The detection was at more than 99% accuracy than an exposed feed. Most importantly, the transformation allows the Navy to retain ownership of its data while using the facial recognition technology. The process ran simultaneously with near real-time recognition without any interference or disruption to the workflow.
The successful deployment of this technology in the U.S. Trident Warrior exercise underscored the significant impact that cutting-edge solutions like Stained Glass Transform™ can have in enhancing security measures, improving operational efficiency, and protecting sensitive information in companies looking to use AI.
Additional Use Cases for Stained Glass in AI
Enabling data ownership and protection in the field of AI and machine learning continues to pave the way for transformative applications within the military and beyond, offering significant benefits to various industries. Here are just a few examples:
1. Security monitoring.
With versatile security monitoring, Stained Glass Transform™ expands far beyond military applications. The program also offers significant enhancements to security monitoring across different industries. For example, at large sporting events and concerts, privacy restrictions pose challenges to surveillance, requiring a solution that could balance real-time threat monitoring with the need to protect sensitive visual data. Similar to the Navy US case, Stained Glass can be implemented in such scenarios.
2. Industrial Automation.
The technology behind Stained Glass Transform™ can extend to diverse domains such as factory floors, where it enables automatic detection of anomalous activity and quality control checks for product items. Companies may want such activities to be monitored while maintaining the confidentiality of proprietary information while inspecting various environments. Stained Glass has helped customers to monitor operations at partner and customer sites, focusing on what to look for while maintaining the privacy customers want.
3. Privacy-enhanced information sharing.
Banks grappling with the task of sharing crucial information to combat fraud face legal restrictions due to personal data present on checks. Fortunately, Stained Glass Transform™ addresses this dilemma by allowing banks to share only the necessary information while obscuring any extraneous or sensitive details. This preserves privacy and security while facilitating collaboration among financial institutions. Check out the Q2 case study.
Ensuring Data Security in AI with Protopia AI
In today’s data-driven landscape, concerns surrounding data collection, privacy, and responsible AI are more pressing than ever. As such, it is imperative for everyone, from small businesses to large conglomerates, to prioritize transparency and adopt privacy-centered solutions. By doing so, organizations can extract valuable insights from data while ensuring the utmost protection of sensitive information.
Protopia AI accelerates the delivery of enterprise adoption by solving the biggest problem in enterprise adoption of AI GenAI: data confidentiality. We’ve gained recognition with in the Gartner Hype Cycle 4 times in 2023 and were a Gartner Cool Vendor in AI Governance and Responsible AI in 2022. More recently, we were selected by AWS in their Generative AI accelerator to help with data ownership and confidentiality in the space.
To explore more real-world use cases, including the U.S. Navy example above, and to learn about Protopia AI’s comprehensive solution offerings, visit the company’s solutions page.
Or, engage with Protopia AI’s experts to learn how to get started with Stained Glass
