12 most innovative launches at RSA 2025

The RSA 2025 conference didn’t just belong to the usual suspects. This year, a new wave of security innovators – some making their debut at RSA – stood shoulder to shoulder with the industry’s biggest names. From deepfake detectors to agentic AI risk maps, the newcomers, alongside seasoned vendors, made a massive impression at the event.

Here are 12 solutions that stood out from the crowd and captured CSOs’ attention with their game-changing innovations.

AppOmni: SaaS security MCP server

AppOmni unveiled the “world’s first” SaaS Security Model Context Protocol server, designed to bring memory-based protection to SaaS environments for the first time. This innovation expands traditional runtime protection beyond endpoints and cloud workloads, enabling real-time threat detection and prevention within SaaS applications. By applying MCP techniques to the SaaS layer, AppOmni aims to close critical visibility and protection gaps, further aligning SaaS security with the standards of modern enterprise defense.

AuditBoard: AI governance solution

AuditBoard announced its AI governance solution, designed to help organizations accelerate AI risk management and promote responsible AI adoption. The solution is designed to streamline AI use case intake, review, and approval processes, establish a centralized repository for approved AI models, and dynamically link AI risks to vendors, assets, and controls for continuous monitoring. By aligning with frameworks like NIST AI Risk Management Framework, AuditBoard aims to enhance oversight, ensure compliance, and scale responsible AI innovation across the enterprise.

Bugcrowd: RTaaS

Bugcrowd introduced its Red Team as a Service (RTaaS), a crowdsourcing offering that connects organizations with a global network of vetted ethical hackers. This service will enable businesses to simulate real-world adversary tactics, techniques, and procedures (TTPs), providing a comprehensive assessment of their security posture. By leveraging the scale of crowdsourcing, Bugcrowd aims to enhance organizations’ preparedness for zero-day attacks and evolving cyber threats.

Cyera: Omni DLP

Cyera unveiled Omni DLP, adding real-time data loss protection (DLP) capabilities to its core data security posture management (DSPM) technology. This marks the largest expansion on its platform to date, enabled by its Trail Security acquisition. Omni DLP aims to offer unified, adaptive data protection across endpoints, networks, cloud, and communication tools. The solution claims to reduce false positives, automate data detection, provide AI-driven governance, and offer a consolidated view of all DLP risks and policies.

DataMinr: Intel Agents

DataMinr announced Intel Agents within its Agentic AI roadmap – autonomous AI capabilities designed to generate critical context for unfolding events, risks, and threats. This innovation aims to enhance real-time decision-making by providing organizations with timely, AI-generated insights for rapidly evolving situations.
 

FlashPoint: Ignite AI upgrades

Flashpoint revealed significant enhancements to its Flashpoint Ignite platform, introducing AI-powered risk discovery, curated threat feeds, asset-centric intelligence, and on-demand data source expansion. These additions aim to make threat intelligence more accessible and actionable, enabling organizations to respond more effectively to emerging threats.

Netarx: Deepfake detection

Netarx introduced its innovative cybersecurity solution designed to combat social engineering attacks, leveraging AI-generated deepfakes. The platform offers real-time detection across voice, video, and email communications, alerting users to potential threats through an on-screen notification. The idea is to safeguard organizations in sectors like finance, healthcare, insurance, and managed service providers from sophisticated impersonation tactics.

Rubrik: Identity Resilience

Rubrik debuted Identity Resilience, a solution designed to secure both human and non-human identities across on-premises, cloud, and SaaS environments. By integrating identity protection into its Zero Trust Data Security platform, Rubrik aims to address the growing threat of identity-based attacks. The solution offers hybrid protection for Active Directory and Entra ID, comprehensive risk analysis to detect dormant or orphaned accounts, and visibility into risky privilege escalations.

SplxAI: Agentic Radar

SplxAI introduced Agentic Radar, an open-source tool designed to provide complete visibility into agentic AI workflows and their vulnerabilities. By mapping dependencies and interactions within AI systems, Agentic Radar is designed to enable CISOs and security teams to proactively identify and mitigate risks associated with complex agentic workflows. The tool integrates with industry frameworks such as OpenAI Agents SDK, CreAI, LangGraph, and n8n, and will be embedded within the SplxAI platform to enhance offensive AI security across development and deployment stages.

X-PHY Inc: Deepfake Detector

X-PHY Inc. introduced its Deepfake Detector, a real-time, on-device solution designed to identify AI-generated media across video, audio, and images. Utilizing multimodal AI, the tool analyzes facial microexpressions, voice patterns, and GAN artifacts to detect manipulations with up to 90% accuracy. Operating offline, it preserves user privacy and integrates seamlessly with platforms like Zoom, Teams, and Chrome.

Huntress: ITDR expansion and new SIEM

Huntress unveiled major updates to its cybersecurity offerings, including enhancements to its Managed Identity Threat Detection and Response (ITDR) solution and a fully managed SIEM platform. The ITDR enhancements include “Unwanted Access” for detecting suspicious login behavior, “Shadow Workflows” to monitor for malicious email inbox rules, along with support for identifying malicious OAuth applications. Additionally, Huntress introduced a fully managed SIEM solution with over 20 new integrations, providing comprehensive visibility and streamlined compliance capabilities.

Abnormal AI: AI agents

Abnormal AI introduced two autonomous AI agents aimed at streamlining employee security awareness training. The AI Phishing Coach offers personalized, interactive training sessions, replacing generic modules with tailored experiences to better engage employees. Complementing this, the AI Data Analyst aims to transform complex security data into actionable intelligence, enabling security teams to swiftly identify and address potential threats.