Tuesday, February 10, 2026
SandboxAQ, a leading innovator in AI-driven cybersecurity, has introduced a new AI-SPM (AI Security Posture Management) offering designed to give enterprises complete visibility into where AI is deployed across their technology environments. The solution evaluates AI assets for vulnerabilities, insecure dependencies, and exposure risks—including prompt injection, unauthorized access, and data leakage. With this launch, SandboxAQ aims to help organizations confront the growing threat of “shadow AI,” which can expose systems to significant risks if left unmanaged.
The company’s recent research underscores the urgency of the issue. Although 79% of organizations currently run AI in production, an overwhelming 72% have never performed a full AI security assessment, and only 6% have adopted a comprehensive AI-native security strategy. Additionally, more than half of surveyed companies reported major concerns about exposed credentials or secrets within their AI systems, yet only 39% have tools in place to safeguard them. These findings arrive at a time when state-sponsored threat actors are reportedly hijacking commercial AI systems to automate cyber-espionage campaigns across corporate and governmental networks. The research highlights a widening industry need for deeper visibility into AI usage and stronger, AI-specific security controls.
AQtive Guard’s AI-SPM solution addresses these challenges by allowing security teams to discover, assess, and secure their entire AI ecosystem—from underlying models to applications and data sources. Unlike traditional security posture tools, which were never designed for AI pipelines or agent-driven systems, SandboxAQ extends its cryptographic scanning capabilities to AI environments. Its deep inspection approach uncovers hidden AI assets and provides a comprehensive “code-to-cloud” understanding of potential threats.
AI Asset Discovery: Automatically identifies models, agents, MCP servers, and all AI components across cloud and code environments.
AI Risk Assessment: Evaluates each asset for exploitable flaws, dependency risks, and exposure to attacks like prompt injection and data leakage.
Policy Enforcement & Compliance: Applies governance frameworks and custom security policies to ensure alignment with internal standards and regulatory requirements.
Continuous Monitoring: Tracks AI workflows in real time to detect anomalies, attacks, or suspicious activity and helps teams manage incidents efficiently.
The AI-SPM offering is currently available to select early-access customers, with wider availability expected in 2026. The company encourages organizations to request access as the cybersecurity landscape becomes increasingly intertwined with AI-driven threats.
Emphasizing the urgency, Jack Hidary, CEO, explained:
“AI is transforming a lot of industries and simultaneously expanding the attack surface faster than traditional security tools can keep up. We’re seeing attackers weaponize AI tools to exfiltrate sensitive data, manipulate internal systems, and automate large-scale intrusions. If organizations don’t have clear visibility into how AI and agents are being used across their environment, they’re operating blindly. Security teams need to act now before an unmanaged AI system becomes the source of their next breach.”
To join our expert panel discussions, reach out to info@intentamplify.com
Recommended News
Tags:
Kustomer expands its AI suite with new Automation and Observability Assistants, giving CX teams deeper visibility and control across workflows, routing, and AI behavior.
HCLTech partners with Dolphin Semiconductor to co-develop low-power, energy-efficient chips, helping firms meet rising performance & efficiency demands.
