Apple has deployed NVIDIA Confidential Computing GPUs on Google Cloud to power inference for its Foundation Models, validating privacy-preserving AI infrastructure at hyperscale. The move directly addresses the security concerns blocking enterprise AI adoption. ITDMs and security architects should treat this as a reference architecture, not a consumer story.
NVIDIA Confidential Computing Powers Apple’s Private Cloud AI Read More »
Descope has released Agentic Identity Hub 2.5, adding granular access policies, autonomous agent support, and human-in-the-loop auth flows for MCP servers and AI agents. The update addresses a growing security gap as enterprises deploy agentic AI using identity patterns designed for humans. ECI Research analyst coverage examines the business risk, developer implications, and competitive landscape.
Descope Agentic Identity Hub 2.5: Securing AI Agents at Scale Read More »
Specops has launched a fixed-price, fixed-scope Active Directory Security Assessment delivered by Outpost24’s CREST-Accredited Offensive Security Team. The engagement maps privilege escalation and lateral movement paths from a low-privileged user to Domain Admin. ECI Research analyst coverage examines the business case, technical depth, and competitive positioning of the offering.
Specops AD Security Assessment: Exposing Attack Paths Before Attackers Do Read More »
Cequence Security is claiming the reference architecture position in AI agent security, pointing to independent convergence from Anthropic, Dr. Chase Cunningham, and the Center for Internet Security around runtime behavioral control over agents. ECI Research breaks down why authentication-first approaches are structurally inadequate for agentic AI, and what ITDMs and developers should do about it now. The MCP governance layer is emerging as the critical battleground for AI agent security investment through 2026.
AI Agent Security: Why Behavior Beats Authentication | ECI Research Read More »
A new Akeyless global survey of 400 IT and security leaders finds that two-thirds of organizations suspect AI agents have already accessed data beyond their intended scope. With a 14-hour average detection window and more than $1 million in annual incident costs, the report exposes a structural gap between how AI agents are provisioned and how enterprises govern them. ECI Research examines what this means for security strategy, developer practices, and the emerging AI agent identity market.
AI Agent Identity Security: The $1M Risk Enterprises Can’t Ignore Read More »
NetFoundry has launched enterprise MCP and LLM gateways that assign cryptographic identities to AI agents and endpoints, closing inbound ports and removing API key distribution from agentic deployments. The announcement addresses a critical security gap as multi-agent AI moves into production at enterprise scale. ECI Research analysis finds this positions NetFoundry at the intersection of zero-trust networking and AI governance, a category without a clear incumbent.
NetFoundry Brings Zero Trust to AI Agents and LLM Gateways Read More »
Palo Alto Networks has integrated Cortex XSIAM with NVIDIA’s DOCA Argus framework, bringing silicon-level, agentless security to NVIDIA AI Factory infrastructure. The partnership also places Prisma AIRS inside NVIDIA’s Enterprise AI Factory Validated Design, establishing a reference architecture for securing agentic AI at scale. ECI Research analysts examine what this means for enterprise security buyers and the teams building on AI infrastructure today.
Palo Alto Networks & NVIDIA Reinvent AI Factory Security Read More »
Willow has launched from stealth with $7 million in seed funding to provide enterprises with visibility and access control over AI agents. The platform, already deployed across 5,000 Wix employees, addresses a governance gap that most enterprise security programs haven’t formally defined yet. ECI Research examines the market dynamics, competitive landscape, and what this means for IT decision-makers and development teams.
Willow Raises $7M to Govern AI Agents in the Enterprise Read More »
Crogl has launched a free, enterprise-grade agentic AI platform for security operations, deployable in minutes with no licensing restrictions. ECI Research examines the product-led growth strategy, the autonomy confidence gap facing agentic security tools, and what ITDMs and security engineers should evaluate before committing to the enterprise tier.
Crogl Launches Free Agentic AI SOC Platform | ECI Research Read More »
HPE has announced production-grade autonomous networking capabilities across its HPE Mist and HPE Aruba Central platforms, claiming the industry’s first fully agentic AIOps networking architecture. New self-driving actions cover capacity optimization, security remediation, and roaming management without human intervention. ECI Research examines the business case, competitive dynamics, and what IT leaders should do next.
HPE Self-Driving Networks: Agentic AIOps Arrives at Scale Read More »