Public-source research
Review what vendors say publicly, the exact quoted source text, the security requirement each statement may support, and what still needs verification.
Showing 341–360 of 1280 evidence records
Microsoft claims Purview data security posture management (DSPM) and data security posture management (DSPM) for AI help discover, secure, and apply compliance controls for AI usage across the enterprise.
discover, secure, and apply compliance controls for AI usage across your enterprise.
Microsoft says Purview-supported AI app categories include other AI apps detected through browser activity and categorized as generative AI in the Defender for Cloud Apps catalog.
Other AI apps that are detected through browser activity and categorized as "Generative AI" in the Defender for Cloud Apps catalog.
Microsoft claims supported AI app interactions can be monitored per user in the tenant and prompts and responses can flow into Purview activity views.
Interactions using supported AI apps can be monitored for each user in your tenant.
Microsoft says Endpoint data loss prevention (DLP) policies can warn or block users from sharing sensitive information with third-party generative AI sites accessed through a browser.
warn or block users from sharing sensitive information with third-party generative AI sites that are accessed via a browser.
Microsoft says Purview-supported AI apps use existing controls so tenant data is not returned or used by an large language model (LLM) when the user lacks access.
ensure that data stored in your tenant is never returned to the user or used by a large language model (LLM) if the user doesn't have access to that data.
Microsoft says Purview Endpoint data loss prevention (DLP) can warn or block browser-based sharing of sensitive information with third-party generative AI sites.
Windows computers that are onboarded to Microsoft Purview can be configured for Endpoint data loss prevention (DLP) policies
Microsoft says Purview Insider Risk Management has a risky AI usage policy template for prompt injection and protected-material access signals.
Use the Risky AI usage policy template to detect risky usage that includes prompt injection attacks and accessing protected materials.
Microsoft says supported AI agents inherit the same Purview security and compliance capabilities as their parent AI app.
Where these AI apps support agents, they inherit the same security and compliance capabilities as their parent AI app.
Microsoft Purview AI protection materials reviewed did not provide a clear Model Context Protocol (MCP), agent-to-agent (A2A), or agent-to-agent security control claim.
Microsoft Purview AI protection materials reviewed did not provide a clear non-human identity or service-account lifecycle security claim.
Palo Alto Networks claims centralized visibility and governance over agents, models, and interactions, including ownership, permissions, and least-privileged access.
Move from pilots to production with full visibility and governance over every agent, model and interaction.
Palo Alto Networks claims automated AI red teaming for applications and agents using simulated real-world threats and security-vulnerability scanning.
AI Red Teaming: Prisma AIRS Provides automated AI Red Teaming that you can use for scanning your AI applications or agents for security vulnerabilities.
Palo Alto Networks claims predeployment model and agent-artifact scanning across models, source code, Model Context Protocol (MCP) servers, and skills.
Scan supply chain vulnerabilities in agent artifacts, including agent code, MCP servers, and skills.
Palo Alto Networks claims centralized control of tool calls, large language model (LLM) interactions, and Model Context Protocol (MCP) connections with granular runtime policy enforcement.
Gain centralized control of tool calls, LLM interactions, and Model Context Protocol (MCP) connections — enforcing granular policies on how agents interact with systems.
Palo Alto Networks claims agent identity inventory and validation, accountable ownership, permissions, least-privileged access, and privilege revocation.
Inventory and validate the identities of agents operating in your enterprise. Define AI agent ownership, permissions and enforce least-privileged access for AI agents.
Palo Alto Networks claims agent artifact and endpoint security covering agent code, Model Context Protocol (MCP) servers, skills, local systems, files, and workflows.
Agentic Endpoint Security extends protection to endpoints, governing how agents interact with local systems, files and workflows.
Cisco claims standards-aligned findings, risk scoring, policy-driven gating, and compliance support across model, agent, and Model Context Protocol (MCP) workflows.
Findings and protections map to leading frameworks, including MITRE ATLAS, OWASP Top 10 for LLMs, and NIST AI-RMF
Cisco claims algorithmic red teaming for models and applications across more than 200 threat subcategories with model-specific guardrail generation.
Automatically test models and applications at AI scale with Cisco algorithmic red teaming.
Cisco claims scanning of model files, repositories, datasets, Model Context Protocol (MCP) servers, and tools for malicious code, poisoning, backdoors, and compromised components before production.
Scan AI model files, repositories, and MCP servers for hidden risks before they are introduced into development or production.
Cisco claims real-time Model Context Protocol (MCP) enforcement, tool allowlists and blocklists, and runtime policies across prompts, responses, agent actions, and tool calls.
MCP runtime enforcement: Monitor and enforce safe behavior across MCP requests and responses connecting LLMs, agents, and tools.