Paul Bryant Most enterprise AI programs still treat agent risk as a prompt problem. That made sense when the agent was little more than a chat interface wrapped around a model. You could improve the system prompt, add better instructions, test a few jailbreak patterns, and feel like you had made the system safer. That model breaks … Explore: The Agent Control Plane: Why Enterprise AI Needs…
Introduction Cloud repatriation has become another architecture debate that generates more heat than evidence. One side treats public cloud as the default destination for every application. The other treats every unexpected bill, provider outage, or jurisdictional concern as proof that workloads should return to privately owned infrastructure. Both positions fail for the same reason: they … Explore: Cloud Repatriation Without Religion: A Workload Placement Engine…
TL;DR A reliable VMware NSX deployment on NVIDIA networking depends less on clicking through the NSX Manager wizard and more on getting the physical underlay right first. The NVIDIA Spectrum fabric must provide stable Layer 3 reachability between every ESXi and NSX Edge tunnel endpoint, consistent jumbo MTU, predictable uplink behavior, and resilient routing to … Explore: How to Install and Configure VMware NSX with…
TL;DR Specification gaming becomes an operational security problem when an autonomous agent can pursue a valid evaluation objective through methods that violate authorization boundaries. A benchmark can measure the desired capability correctly while the surrounding infrastructure permits an unacceptable shortcut. The required response is not a longer system prompt. High-capability evaluations need independently enforced invariants … Explore: When Benchmark Cheating Becomes a Production Breach: Specification…
TL;DR Security teams need AI systems that can inspect the material most general-purpose assistants are designed to treat cautiously: exploit code, malware behavior, command-and-control traffic, exposed credentials, persistence mechanisms, and destructive commands. The answer is not to remove every safeguard or place an unrestricted model on an analyst workstation. The safer pattern is a governed … Explore: The Guardrail Paradox: Designing a Governed Forensic AI…
TL;DR Traditional logs tell operators what individual components recorded. Agent observability must answer a harder question: did an autonomous system remain inside its declared task, authorization boundaries, and approved methods across the complete sequence of actions? The required unit of detection is the trajectory. Prompts, tool calls, shell commands, identities, network destinations, package retrieval, privilege … Explore: Agent Observability Is Not Logging: How to Detect…
TL;DR The OpenAI and Hugging Face incident should not be interpreted as a model spontaneously acquiring authority or developing independent malicious intent. According to OpenAI’s preliminary disclosure, models running inside a cyber evaluation were given a narrow objective, reduced behavioral refusals, substantial compute, a sandbox with package access, and an infrastructure path that could be … Explore: The Agent Was Not the Rogue: OpenAI’s Cyber…
TL;DR An autonomous AI agent with shell access, credentials, package installation, tools, and network connectivity should be treated as a potentially hostile privileged workload. The security boundary cannot depend on the task description, system prompt, or assumption that the agent will follow the expected execution path. Controls must be based on the maximum authority the … Explore: Your AI Agent Is a Privileged Insider: A…
Introduction GPU sharing is easy to describe and difficult to govern. A platform team can expose one physical accelerator as several scheduler-visible resources, divide it into Multi-Instance GPU partitions, present virtual GPUs to virtual machines, or assign the entire device to one workload. Kubernetes can place those workloads into separate namespaces. An enterprise scheduler can … Explore: GPU Multi-Tenancy Without Security Theater: Isolation, Quotas, Noisy…
Introduction A shared GPU platform creates a deceptively simple question: who used the GPU? The question becomes difficult as soon as the platform supports more than one operating model. A Kubernetes pod may receive an entire GPU, a MIG instance, or a time-sliced share. A virtual machine may receive a vGPU profile or a pass-through … Explore: Who Used the GPU? Building Per Tenant Telemetry,…
Introduction An enterprise inference service can look busy while its GPU remains nearly idle. The application may be accepting requests, retrieving documents, validating permissions, tokenizing prompts, waiting on storage, retrying dependencies, or building responses. None of those activities prove that enough executable work is reaching the accelerator. This is why GPU troubleshooting often goes wrong. … Explore: Why Your GPU Is Idle: A Layer by…
TL;DR MCP and A2A solve different integration problems and belong at different layers of the enterprise agent architecture. Use MCP when an agent needs controlled access to tools, resources, APIs, data, and bounded operations. Use A2A when one independently operated agent needs to discover another agent, delegate a goal, participate in a multi-turn interaction, track … Explore: How MCP and A2A Fit into One Enterprise…
TL;DR An agent tool is not merely an API endpoint with a JSON wrapper. It is a contract between a nondeterministic decision-maker and a deterministic system. Reliable tools have distinct names, narrow responsibilities, constrained input schemas, useful descriptions, predictable output structures, retry-safe side effects, actionable errors, server-side validation, and evaluations built around realistic tasks. The … Explore: How to Design Tools That AI Agents Can…
Introduction The most common sizing question about VMware Cloud Foundation is still framed as a host-count question: How many ESX hosts are required before VCF can be deployed? That question is useful for checking a deployment prerequisite, but it is not sufficient for an architecture decision. A platform can satisfy a minimum host count and … Explore: The Minimum Viable VMware Cloud Foundation 9.1: Hosts,…
TL;DR Enterprise retrieval-augmented generation should be designed as an evidence system, not as a chatbot with a vector database attached. The reliable pattern is to ingest authoritative content with stable metadata, enforce access control before retrieval, combine keyword and vector search, rerank a larger candidate set, package the surviving chunks as explicit evidence, and require … Explore: How to Build Enterprise RAG That Returns Evidence,…
TL;DR A GPU-backed workload is not recoverable merely because its virtual disks were copied or its Kubernetes manifests were committed to Git. vGPU, MIG-backed vGPU, PCI passthrough, Enhanced DirectPath, RDMA, and bare-metal GPU configurations expose different snapshot, suspend, migration, and failover boundaries. Some support controlled mobility within a narrow compatibility matrix. Others deliberately trade those … Explore: How Do You Protect a GPU-Backed Workload When…
Human approval is not useful just because a workflow pauses. It becomes useful when the system can prove what was proposed, why review was required, who reviewed it, what evidence they saw, what scope they approved, what executed, and whether the outcome matched the approved intent. That is the difference between an approval button and … Explore: From Approval Button to Control Plane: Implementing Human…
TL;DR A private GPU hour is not the purchase price of a server divided by the number of GPUs and hours in a year. A defensible enterprise AI FinOps model must include depreciation, software, support, power, cooling, facilities, network, storage, platform labor, maintenance, failure reserves, and the difference between installed, schedulable, allocated, and productive GPU … Explore: What Does a Private GPU Hour Actually Cost?…
Introduction Enterprise GPU design becomes confused when several different decisions are compressed into one question: “How should we share the GPU?” That question mixes hardware assignment, virtualization, Kubernetes scheduling, tenant isolation, business priority, and service-level commitments. The result is often a platform that advertises many GPU “slices” but cannot explain what each slice guarantees. A … Explore: Whole GPU, Passthrough, vGPU, MIG, or Time Slicing?…
VCF 9.1 creates a familiar operational trap: a platform team can complete a release-note review and still not be ready to upgrade. The reason is that support changes do not arrive as one clean class of event. Some capabilities remain functional but are deprecated. Some hardware operates under restricted support. Some components have already been … Explore: VCF Deprecations and Removals: An Upgrade Readiness Guide
