Space and defense mission systems increasingly rely on complex software ecosystems and AI-enabled capabilities operating in contested and resource-constrained environments. These conditions introduce challenges in software assurance, supply chain visibility, and the reliable deployment of AI within operational systems. At the 8th Annual SSC Cyber Expo, ObjectSecurity will exhibit BinLens and FortiLayer, demonstrating approaches to strengthening assurance across both traditional software and AI-enabled mission components. The focus is on operationally relevant capabilities that address risks present in deployed and fielded systems.
As mission systems increasingly rely on complex and third-party software, teams are being asked to trust binaries they did not build and cannot fully inspect.
In this session, ObjectSecurity presents BinLens, an automated binary vulnerability analysis platform designed to uncover security flaws directly in compiled software. BinLens analyzes binaries to expose weaknesses that are missed by source-based tools, SBOM checks, and signature-driven scanners, including unknown, potential zero-day vulnerabilities.
Attendees will see how BinLens helps teams:
- Detect zero-day and unpublished vulnerabilities in compiled binaries
- Reduce inherited risk from vendors and third-party software
- Generate actionable findings with low false positives for engineering and security teams
If you’re responsible for approving, deploying, or defending mission software you didn’t write, this session will show how to surface real risk before it becomes an operational problem.
FortiLayer is ObjectSecurity’s platform for securing AI-enabled systems across their lifecycle, with emphasis on runtime assurance in deployed environments. It provides visibility into model behavior, enabling detection of anomalies, validation of performance, and continuous monitoring of system integrity.
As AI becomes embedded in mission systems, large language models introduce additional risks related to prompt manipulation, behavioral drift, and unreliable outputs. Conventional monitoring approaches focus on prompts and responses, but do not provide insight into how the model arrives at those outputs.
FortiLayer extends analysis beyond inputs and outputs by exposing internal model behavior during execution. This enables identification of abnormal or adversarial influence as it occurs, rather than relying solely on post hoc evaluation.
With this approach, FortiLayer supports:
- Detection of adversarial prompt influence and abnormal model behavior
- Analysis of model response generation under dynamic conditions
- Identification of hidden failure modes and unintended data exposure risks
- Continuous evaluation of AI system integrity during operation
FortiLayer is designed for use in operational environments, including systems with resource constraints and real-time requirements. It supports deployment scenarios where understanding model behavior is necessary to maintain trust and ensure mission reliability.
This work reflects ongoing efforts to provide actionable assurance for AI systems, including LLM-based capabilities, operating in mission-relevant and adversarial conditions.
