Zenity

Gartner doesn't just evaluate what a vendor builds, in our view, it evaluates whether they can execute. 🚀 Zenity was just named a Market Shaper in the September 2026 Gartner® Emerging Market Quadrant for AI Application Security - Startup Vendors, one of only two vendors placed there among nearly 20 assessed. This is another recognition from Gartner in 2026, following Zenity being named as the Company to Beat in AI Agent Governance. We believe these two reports cover the most critical dimensions of the emerging AI security market. 👉 https://lnkd.in/eCms3n4G  #AISecurity #AgenticAI #Cybersecurity #AIAgentSecurity

Dionisio Zumerle is one of the lead authors on the recently released eMQ where Zenity is recognized as a Market Shaper, sitting in the upper right hand quadrant. Dionisio is presenting on this research, helping users make sense of one of the fastest moving spaces. Thank you Dionisio for bringing clarity here to attendees at the Gartner SRM London event held earlier this week. 🎉 https://lnkd.in/ejmW4aaV #AISecurity #AgenticAI #Cybersecurity Gartner

An agent's permissions can look correct on paper while its real behavior drifts from what anyone approved. That drift is where the risk actually lives. ⚠️ September 2026 Gartner® Emerging Market Quadrant for AI Application Security - Startup Vendors just named Zenity a Market Shaper. Zenity provides continuous, event-driven posture management and adaptive testing that catches privilege escalation and data exfiltration at runtime, not after the fact. This is another recognition from Gartner, following Zenity being named as the Company to Beat in AI Agent Governance in April 2026. These two reports cover the most critical dimensions of the emerging AI security market. We break down what we feel this Emerging Magic Quadrant covers and why it matters for teams running agents in production. 👉 https://lnkd.in/ejmW4aaV #AgenticAI #AISecurity #AIGovernance #AIAgentSecurity Gartner

Zenity has been named a Market Shaper in the September 2026 Gartner® Emerging Market Quadrant for AI Application Security - Startup Vendors. 🎉 Most AI application security tools stop at the mandatory baseline: scan the code, confirm a control exists, move on. AI applications introduce risks such as prompt injection, sensitive data leakage, and rogue agent action. Startups in AI application security offer solutions to secure and protect custom-built AI, AI agents, and models. Cybersecurity leaders can use these insights to evaluate startups in this new market evolution. This recognition comes shortly after Zenity was named the Company to Beat in AI Agent Governance. We believe these two reports cover the most critical dimensions of the emerging AI security market. Read more 👉 https://lnkd.in/e6M9niri #AISecurity #AgenticAI #AIAgentSecurity #Cybersecurity Gartner

Agentic security failures aren't hypothetical. Join Zenity Field CISO, Gary Hayslip at the AI Agent Security Summit in NYC on October 21, where we'll be breaking down where agents have gone rogue in 2026, and what's on the horizon. See the full agenda and register here: https://lnkd.in/emcRQ_Ku #Zenity #AIAgents #AISecurity #Cybersecurity

Most "shadow AI" isn't rogue behavior. It's your best employees moving fast. 🕵️ A sales rep connects an AI note-taker to their calendar. An analyst builds a small automation inside the CRM to draft outreach. No one calls it "building an agent." It's just getting the job done. That's exactly why shadow AI is so hard to catch. It doesn't look unauthorized because most of the time, it isn't malicious. It's productivity, running through a tool nobody reviewed. Here's the harder problem: finding these agents isn't the finish line. An agent that's discovered and then sits in a queue has only changed from invisible to unaddressed. Knowing an agent exists doesn't tell you whether its access is appropriate, or what it's already done with the access it has. Traditional shadow IT tooling was built to catch unauthorized infrastructure. Shadow AI lives somewhere different: inside the prompt, inside the agent's configuration, inside the actions it's authorized to take once it's connected to a live system. Closing that gap takes more than visibility. It takes continuous, context-aware judgment on what every agent is doing right now, not just what it was technically allowed to do. 👉 Read the full breakdown: https://lnkd.in/eUQmvxh5 #AI #AISecurity #AgenticAI #Cybersecurity #ShadowAI

Healthcare is adopting agentic AI faster than almost any other regulated industry. 🏥 Ambient scribes documenting visits in real time. Prior-authorization agents working payer workflows. Agents reading and writing straight to the EHR through FHIR and MCP connections. The value is real. So is the gap. CMS's Prior Authorization API mandate takes effect January 1, 2027. The HIPAA Security Rule update isn't expected until mid-2027. FDA guidance on AI-enabled clinical decision support only covers tools where a clinician can independently review the recommendation. None of these mandates were written with autonomous agents in mind. They move data. They don't define what an agent is allowed to do once it's connected. An agent that was only ever supposed to check a lab result but technically can write to a medication list is a governance gap, whether or not it's ever misused that path. Detection after a record has already moved isn't security. Intent isn't control either. The organizations pulling ahead aren't waiting for the next rulemaking cycle. They're building agent-level visibility and runtime controls now, so they can keep expanding into sensitive workflows with evidence, not assumptions. 👉 Read the full breakdown by Robert Clinton: https://lnkd.in/e_yGJaNk #AISecurity #AgenticAI #Healthcare #AIAgentSecurity #Cybersecurity

Non-human identities now outnumber humans 82 to 1 in the average enterprise. 🎯 AI agents are the fastest-growing category inside that number, and they break the access model everyone built for service accounts. Here's why: a service account uses the access it's given. An agent reasons about how to use it, chains it across tools, and decides what to touch next without a human in the loop. That's why the PocketOS incident is so unsettling. No attacker. No stolen credentials. An agent held standing permissions it technically had, reasoned its way into the production database, and lost the entire thing in nine seconds. Every access check passed. It was still catastrophic. Static, quarterly access reviews can't catch an agent whose job changes task by task. What agents need is runtime least privilege: scoped to the current step, granted right before it's needed, revoked the moment it's done. Our new Academy article breaks down what that looks like in practice, and the governance controls IAM teams need to build toward. 👉 Read the full guide: https://lnkd.in/eEbEEMhb #AgenticAI #AISecurity #AIAgentSecurity #Cybersecurity #EnterpriseAI