Anaconda's Enkrypt AI featured in two Gartner reports
According to Anaconda, Enkrypt AI is mentioned as a Sample Vendor for AI security testing in two 2026 Gartner Hype Cycle reports. The documents examine AI-driven cybersecurity and secure AI.
- Enkrypt AI is listed as a Sample Vendor in two 2026 Gartner Hype Cycle reports.
- The service tests models, AI agents and MCP connections using attack scenarios.
- According to Anaconda, more than 268,000 tools and 25,000 MCP servers were assessed over two months.
- Alongside testing, the platform provides runtime guardrails and compliance evidence.
Included in two Gartner reports published in October
Enkrypt AI, which is now part of Anaconda, is listed as a Sample Vendor in the AI Security Testing category in Gartner's Hype Cycle for AI-Driven Cybersecurity, 2026 and Hype Cycle for Secure AI, 2026 reports. The reports were published on October 12, 2026.
Gartner defines this category as covering the discovery of vulnerabilities and exposures in systems and applications that use artificial intelligence. This may include automated adversarial testing, such as generating and executing adversarial prompts, as well as assessing models, libraries, frameworks and notebooks.
According to Anaconda, Enkrypt AI focuses on the adversarial side of the definition. The service tests models, AI agents and the connections they use before release, then uses the test results to develop guardrails applied during operation.
Testing examines the agents' entire operating chain
The company cites the example of a hospitality customer service chatbot. During red teaming, the system confirmed a nonexistent booking, recommended a competitor's property instead of its own, and quoted different prices for the same room based on demographic cues in the conversation. According to Anaconda, none of these incidents involved a jailbreak, yet the chatbot created brand, legal and revenue risks.
The tests also examine how ingested content influences an agent's decisions, which tool it calls and what action the tool performs. For example, an instruction hidden in an uploaded PDF could cause an agent to initiate an unrequested refund.
According to Enkrypt AI, red teaming simulates text-, image- and audio-based attack scenarios. The company says it has assessed more than 268,000 tools across 25,000 Model Context Protocol, or MCP, servers over the past two months and found more than 143,000 vulnerabilities. These issues affected 73 percent of the servers.
The platform spans pre-release testing through compliance
According to Anaconda, Agent Red Teaming assesses agents, models and MCP connections across more than 300 attack categories, with more than 50,000 dynamic evaluations per model. MCP Scanner checks the MCP servers used by agents, while the Anaconda Platform's AI security and guardrails features continuously examine packages and models for issues including security vulnerabilities, licensing problems and malicious code.
Agent Guardrails protects against the failure modes identified by red teaming during operation. Agent Policy Engine generates auditable evidence aligned with the European Union AI Act and the NIST AI Risk Management Framework. According to the company, agents can be retested after every change to a model, prompt or tool.
Anaconda's previous Gartner mentions include being named a Cool Vendor in AI Security in September 2025, as well as receiving the Specialist designation in the AI Application Security, Startup Vendors category in September 2026. In October 2026, Enkrypt AI was also listed as a Sample Vendor for multimodal AI protection. Gartner states, however, that its publications do not constitute endorsements of products or companies.
