Calypsoai Named A Top-Two Finalist At RSACTM 2025 Innovation Sandbox, Defining The Future Of AI Security
CalypsoAI's groundbreaking Inference Platform offers a full-lifecycle AI security architecture that protects AI systems and agents at inference - where decisions are made, risks emerge, and trust must be earned. The panel of industry expert judges praised CalypsoAI's focus on securing AI systems and agents, which have the potential to transform enterprises in the immediate future.
The RSA Innovation Sandbox is renowned for spotlighting transformational companies that redefine the security landscape. Finalists have collectively raised over $16.4 billion and driven over 90 acquisitions since the contest's inception.
More than 200 companies applied for this year's Innovation Sandbox contest, which is in its 20th year. The judges selected CalypsoAI as a top-two finalist from ten finalists, following product demonstrations, a live onstage pitch, and a Q&A session at RSACTM 2025 at the Moscone Center in San Francisco.
CalypsoAI's platform stands apart by delivering:
-
The CASI Score (CalypsoAI Security Index): A proprietary metric quantifying security beyond simplistic attack success rates, offering enterprises actionable transparency and continuously updated benchmarks for AI security posture.
Agentic WarfareTM and Resistance Scoring: Deploying a team of autonomous agents to evaluate and continuously red-team AI systems, exposing vulnerabilities introduced by model choice, system architecture, and data access.
AI Security Gateway with Deep Content Inspection: Blocking prompt injection, data leakage, and adversarial misuse in real time.
Autonomous Exposure Management: Combining agentic-powered red-teaming and real-time defense to identify and remediate vulnerabilities before they cause harm.
CalypsoAI's Inference Platform currently runs over 50,000 continuously evolving signature attacks , delivering the most advanced real-time protection. Recent stress tests saw CalypsoAI's Inference Red-Team agents successfully compromise every major foundation model, proving the urgent need for active protection.
As AI systems become more autonomous and the threat landscape more complex, enterprises need security platforms that can adapt and evolve with them. CalypsoAI's Platform is built to scale alongside teams - providing flexible, modular defenses that grow as AI use cases expand. By continuously learning from adversarial testing and threat evolution, CalypsoAI ensures enterprises can stay secure today and resilient tomorrow.
"CalypsoAI's agents are unlike anything else on the market - adversarial, autonomous, and built to mirror the evolving threats enterprises face," added James White, President and CTO of CalypsoAI. "By combining agentic red-teaming, real-time inference defense, and continuous observability, we've created a self-healing security architecture that ensures AI stays resilient as it scales."
Following this achievement at RSAC, CalypsoAI plans to accelerate deployment of its Inference Platform - expanding real-time protection, adaptive governance, automated remediation, and dynamic exposure management for enterprises worldwide that are deploying AI.
About CalypsoAI
CalypsoAI secures AI systems and agents through its Inference Platform - protecting AI with customizable defensive controls, agentic red-teaming, governance, observability, and coming soon, automated remediation. Trusted by organizations like Palantir and SGK, and backed by investors including Paladin Capital Group, Lockheed Martin Ventures, and Hakluyt Capital, CalypsoAI ensures enterprises can innovate with AI safely, responsibly, and at scale.
Learn more at calypsoai and follow us on LinkedIn and X .
Media Contact:
Louise Scully, Head of Marketing
[email protected]
Photo: g
Logo: g
SOURCE CalypsoAI
Legal Disclaimer:
MENAFN provides the
information “as is” without warranty of any kind. We do not accept
any responsibility or liability for the accuracy, content, images,
videos, licenses, completeness, legality, or reliability of the information
contained in this article. If you have any complaints or copyright
issues related to this article, kindly contact the provider above.

Comments
No comment