Enkrypt AI, now part of Anaconda, has been named a Sample Vendor for AI Security Testing in two Gartner® Hype Cycle™ reports published in October 2026:
- Hype Cycle for AI-Driven Cybersecurity, 2026
- Hype Cycle for Secure AI, 2026
In our experience, AI agents and AI applications introduce risks that teams don’t always recognize as security issues.
For example, when we red teamed a customer-facing hospitality chatbot, it confirmed a reservation that didn’t exist, recommended a competitor’s property over its own, and quoted different prices for the same room depending on demographic signals in the conversation. None of these were jailbreaks. The bot did what users asked, in ways that created brand, legal, and revenue risk. Those are the failures adaptive red teaming is built to find.
What AI security testing covers
In our experience, AI agents change faster than security reviews can keep up, so AI agent and AI application security testing has to run continuously.
Both Hype Cycles define the category as follows:
“AI security testing (AI-ST) uncovers vulnerabilities and exposures in AI-enabled systems and applications. It offers specialized assessments for AI applications, including offensive tactics like automated generation and execution of adversarial prompts and AI component scanning — model repositories, libraries, frameworks and notebooks. AI-ST can function independently, integrated in AI security or application security tools.”
Enkrypt AI focuses on the adversarial side of that definition: automated attacks that probe models, agents, and the tools they connect to before release, with results that feed the guardrails running in production. Testing without runtime defense gives you a report and no protection.
What testing finds in practice
The most damaging failures we see play out over several steps. Instructions in a retrieved document can change an agent’s plan, which determines the tool it calls, and the tool takes an action outside the agent’s approved scope. For example, an instruction hidden in an uploaded PDF can get an agent to issue an unrequested refund. Our red teaming tests each link in that chain:
- What the agent reads: Our red teaming simulates adversarial scenarios across vision, text, and audio, because a poisoned screenshot can redirect an agent’s tools and a spoofed voice can trigger approvals.
- How the model decides: We have found exploitable risks across frontier AI providers and open source models, and models from different providers respond very differently to the same attacks. That is why we maintain the LLM Safety Leaderboard, which benchmarks 300 models.
- What the agent can do: In two months, Enkrypt AI scanned 268,000+ tools across 25,000 Model Context Protocol (MCP) servers, uncovering 143,000+ vulnerabilities affecting 73% of those servers.
Where Enkrypt AI fits
Our customers tell us Enkrypt AI gives them confidence that their AI-powered chatbots and AI agents provide safe and compliant answers to their customers and employees. That confidence starts with testing before release and continues with protection in production.
| What the job requires | How Enkrypt AI by Anaconda handles it |
|---|---|
| Attack models and agents before release | Agent Red Teaming autonomously red teams models, agents, and MCP connections across 300+ attack categories, with 50,000+ dynamic evaluations per model. |
| Test the tools agents use | MCP Scanner scans the MCP servers agents connect to for risk. |
| Test inputs beyond text | Red teaming uses adversarial image-text pairs, visual perturbations, and audio distortions for multimodal systems. |
| Scan packages and models | AI security & guardrails in the Anaconda Platform scans packages and models continuously, catching security vulnerabilities, license issues, and malicious code before production. |
| Turn findings into protection | Agent Guardrails stop the failure modes red teaming discovers once the agent is running. |
| Prove controls work | Agent Policy Engine generates audit-ready evidence mapped to the EU AI Act and the NIST AI Risk Management Framework. |
For the hospitality chatbot above, Enkrypt AI red teaming surfaced the fake reservation, the competitor recommendation, and the price differences. Findings like these show a team which guardrails and policies the bot needs and Enkrypt AI can apply them for production.
Together, these capabilities help security teams find risks before an AI agent ships, retest as it changes, provide runtime defense in production, and show compliance reviewers which policies were set and enforced in production.
See how Enkrypt AI tests agents before release.
Enkrypt AI’s recent Gartner recognitions:
- September 2025: Named a Gartner Cool Vendor in AI Security.
- September 2026: Named a Specialist in the first Gartner Emerging Market Quadrant for AI Application Security — Startup Vendors.
- October 2026: Named a Sample Vendor for Multimodal AI Protection in the Gartner Emerging Tech Impact Radar: AI Cybersecurity Ecosystem.
- October 2026: Named a Sample Vendor for AI Security Testing in the Gartner Hype Cycle for AI-Driven Cybersecurity, 2026 and the Hype Cycle for Secure AI, 2026.
Where to start
Pick one agent on this quarter’s release list and red team it before it ships. Then retest it each time its model, prompts, or tools change. The results show which guardrails it needs and give your security and compliance reviewers concrete evidence to sign off on.
See Enkrypt AI in action. Request a demo with one of your own agents and see what red teaming finds and what guardrails stop.
Gartner clients can read the full Hype Cycle for AI-Driven Cybersecurity, 2026 and Hype Cycle for Secure AI, 2026 on gartner.com.
FAQs
What is a Gartner Hype Cycle?
In Gartner’s words, “Gartner Hype Cycles provide a graphic representation of the maturity and adoption of technologies and applications, and how they are potentially relevant to solving real business problems and exploiting new opportunities.”
What did Gartner name Enkrypt AI by Anaconda?
In the Hype Cycle for AI-Driven Cybersecurity, 2026 and the Hype Cycle for Secure AI, 2026, both published October 12, 2026, Gartner named Enkrypt AI a Sample Vendor for AI Security Testing.
Has Gartner recognized Enkrypt AI before?
Yes. Gartner named Enkrypt AI a Cool Vendor in AI Security in September 2025, a Specialist in the Emerging Market Quadrant for AI Application Security — Startup Vendors in September 2026, and a Sample Vendor for Multimodal AI Protection in the Emerging Tech Impact Radar: AI Cybersecurity Ecosystem in October 2026.
Why do AI agents and AI applications need their own security testing?
Traditional application security assumes code behaves the same way on every run. AI agents and applications respond differently depending on prompts, retrieved content, and tool responses, so certain risks only appear at runtime. Agents raise the stakes because they act on those inputs.
How does Enkrypt AI relate to Anaconda?
Enkrypt AI is part of Anaconda. Enkrypt AI by Anaconda delivers AI security & guardrails in the Anaconda Platform, one of four platform pillars alongside AI workspaces, AI orchestration, and AI artifacts.
Gartner, Hype Cycle for AI-Driven Cybersecurity, 2026, Jeremy D’Hoinne, Craig Porter, Josh Murphy, Christine Lee, 12 October 2026, ID G00854093.
Gartner, Hype Cycle for Secure AI, 2026, Jeremy D’Hoinne, Dionisio Zumerle, Jeff Arias, 12 October 2026, ID G00846820.
Gartner, Emerging Tech Impact Radar: AI Cybersecurity Ecosystem, David Senf, Mark Wah, Tarun Rohilla, Marissa Schmidt, Evan Zeng, Alfredo Ramirez IV, Tuong Nguyen, Walker Black, 5 October 2026, ID G00846953.
Gartner, Emerging Market Quadrant for AI Application Security — Startup Vendors, Meghan Hollis, Dionisio Zumerle, Dennis Xu, Marissa Schmidt, 16 September 2026, ID G00856640.
Gartner, Cool Vendors in AI Security, Jeremy D’Hoinne, Bart Willemsen, Dennis Xu, Avivah Litan, 24 September 2025.
GARTNER and HYPE CYCLE are trademarks of Gartner, Inc. and/or its affiliates.Gartner does not endorse any company, vendor, product or service depicted in its publications, and does not advise technology users to select only those vendors with the highest ratings or other designation. Gartner publications consist of the opinions of Gartner’s business and technology insights organization and should not be construed as statements of fact. Gartner disclaims all warranties, expressed or implied, with respect to this publication, including any warranties of merchantability or fitness for a particular purpose.