News for today:
- [Silverfort] Integration with Copilot Studio. Silverfort integrated its Identity Security platform for AI agents with Microsoft Copilot Studio, for enforcing identity security whenever a Copilot agent runs. Silverfort integrates into the Copilot Studio ecosystem to enable runtime enforcement. Every time a Copilot agent requests access to a tool or function, Silverfort evaluates the request in real time and returns a decision before the action executes. This prevents unauthorized access, privilege escalation, and unintended actions before they can occur, allowing organizations to adopt AI agents with confidence. Silverfort
- [Wallarm] GA of Infrastructure Discovery. Wallarm announced GA of its new Infrastructure Discovery capability on AWS; it’s part of the recently introduced AI Control Platform. Infrastructure Discovery continuously maps every AWS account, region, service, AI workload, and agent in your environment, so security has a complete inventory before anything reaches production. AI is the new business operating system, and Infrastructure Discovery uncovers and secures the foundation. There’s a free usage tier as well as others. Wallarm
- [Descope] v2.5 of Agentic Identity Hub. Descope released v2.5 of its Agentic Identity Hub, with new capabilities for granular authorization rules, support for autonomous agents, standalone MCP authorizations, and more. Organizations will now be able to use Descope to manage identities for headless agents, enable scope-based agent access to backend APIs, implement step-up auth flows for sensitive agent actions, and become agent-ready without changing existing user auth systems. Descope also announced its MCP server to let developers use AI agents to run authentication, analytics, and migration operations on their Descope projects. Descope
- [Tenable] AI-powered cloud detection and response. Tenable released new AI-powered cloud threat detection capabilities in its exposure management platform, for correlating alerts and prioritizing focus areas for security teams. Tenable bridges the gap between cloud threat activity and unified risk visibility for proactive exposure management. Tenable goes beyond static misconfigurations and vulnerabilities, contextualizing runtime telemetry within the broader attack surface, empowering security teams to stop chasing theoretical risks and prioritize remediation based on true business impact. Tenable
- [Saviynt] Expanded partnership with Zscaler. Saviynt and Zscaler expanded their partnership, aligning identity-driven decisions with real-time enforcement. The initial integration focuses on eliminating standing privileges through just-in-time (JIT) access. This addresses enterprises still reliant on fragmented access models tickets, VPNs, group memberships, and persistent credentials that extend privileges beyond their intended use. Saviynt governs identity, validates access requests, and provisions time-bound entitlements, while Zscaler enforces those decisions inline at session initiation through the Zscaler Zero Trust Exchange platform. Saviynt
- [F5] New web application and API protection. F5 added new capabilities to its platform for protecting web applications and APIs. Intent is to provide better protection against the weaponization of frontier AI models. F5
- [CrowdStrike] Activities of China-nexus adversaries. New research from CrowdStrike shows that China-nexus adversaries are targeting technology companies to steal AI capabilities and related IP. Other findings: DPRK-nexus adversaries are accelerating fraudulent IT worker schemes to funnel revenue to the regime, while eCrime actors are weaponizing AI and turning the developer ecosystems behind it into attack vectors. CrowdStrike
- [Fastly] AI traffic growing 6.5X faster than human traffic. New research from Fastly notes that AI requests grew around 30% between January and May 2026, or 6.5 times faster than human requests during the same timeframe. The findings suggest that AI traffic management is evolving from a security and infrastructure concern into a broader business strategy. As organizations evaluate how AI systems interact with their content, applications, and APIs, visibility and control are becoming increasingly important. Fastly