Enterprise Technology Glossary

5,975 results · page 8 of 299

• Agent Policy Engine

  Agent policy engine is a software component that evaluates and enforces formal policies on the behavior and permissions of autonomous or semi-autonomous agents, enabling centralized control, governance, and auditability of agent actions in enterprise and distributed technical environments.

• Agent Reasoning Graph

  Agent reasoning graph is a graph-structured representation of an AI agent’s step-by-step decision process, used in enterprises to trace intermediate reasoning states, support auditability and observability, and integrate AI agent behavior with governance, monitoring, and architectural controls.

• Agent Registry

  Agent registry is a governed repository that stores and describes software-based agents, including AI and autonomous components, so enterprises can discover them, control their lifecycle, and enforce security and governance policies across distributed or multi-agent systems.

• Agent Runtime Environment

  Agent runtime environment is the execution and control layer that hosts and manages autonomous or semi-autonomous software agents in enterprise systems, providing resources, lifecycle management, communication, and governance so agents can operate, coordinate, and integrate with applications, data platforms, and security controls.

• Agent Swarm Network

  Agent Swarm Network is a distributed multi-agent system pattern in which autonomous software agents coordinate via network communication without centralized control. It matters for enterprises that design decentralized robotics, sensing, and AI workloads that require coordinated behavior across many nodes.

• Agent Telemetry Collector

  Agent telemetry collector is a software component that receives and aggregates telemetry from instrumentation agents and routes it to monitoring, observability, or analytics platforms, enabling centralized data collection, cost control, and governance across distributed enterprise infrastructure and applications.

• Aggregated Query Framework

  Aggregated Query Framework currently lacks a stable, source-backed definition in academic, standards, and enterprise research, so it is not a recognized term of art in data or analytics architecture and requires local clarification wherever it is used.

• Aggregation Layer

  Aggregation layer is an architectural tier that consolidates network traffic, data, or service calls from multiple sources into fewer downstream channels, allowing enterprises to centralize policy enforcement, standardization, and observability while separating access or edge components from core systems and services.

• AI-Accelerated Supercomputing

  AI-accelerated supercomputing is a high-performance computing approach that combines traditional supercomputer architectures with specialized AI accelerators and software, enabling enterprises to execute large-scale simulation, data analytics, and machine learning workloads that exceed the performance of general-purpose compute environments.

• AI Acceleration

  AI acceleration is the use of specialized hardware, software, and architectures to execute artificial intelligence workloads with higher computational efficiency, helping enterprises meet performance, latency, and energy requirements for training and inference within data center, cloud, and edge environments.

• AI Accelerator

  AI accelerator is a specialized hardware processor for executing artificial intelligence and machine learning workloads more efficiently than general-purpose CPUs, relevant to enterprises that need higher performance and energy efficiency for model training, inference, and large-scale data-driven applications.

• AI Accelerator Chip

  AI accelerator chip is a specialized integrated circuit that executes artificial intelligence and machine learning workloads with architectures optimized for parallel numeric computation. It matters in enterprises because it underpins performance, efficiency, and capacity planning for AI training and inference deployments.

• AI Accelerator Module

  AI accelerator module is a hardware component or pluggable unit that executes AI workloads, such as neural network inference and training, more efficiently than general-purpose CPUs, enabling enterprises to meet latency, throughput, power, and cost constraints in AI deployments.

• AI access control

  AI access control is the set of enterprise policies and enforcement mechanisms that regulate who and what can access AI models, data, and endpoints, and under which conditions, to support security, compliance, and governed use of AI capabilities.

• AI Accountability Report

  AI accountability report is a structured governance document that records how an AI system is designed, tested, monitored, and controlled, enabling enterprises to demonstrate compliance with internal policies, risk frameworks, and external regulatory or audit requirements for automated and AI-enabled systems.

• AI Agents

  AI agents are software entities that apply artificial intelligence methods to perceive inputs, maintain state, and autonomously execute actions toward specified goals under constraints, which matters in enterprises for controlled automation, decision execution, and integration of AI into operational workflows.

• AI Alignment

  AI alignment is the discipline of designing, governing, and monitoring AI systems so their objectives, constraints, and behaviors match defined human and organizational goals, enabling reliable, policy-compliant use of AI within enterprise architectures and formal risk, security, and compliance frameworks.

• AI Alignment Benchmark

  AI alignment benchmarks are evaluation protocols, datasets, and scoring methods that enterprises use to measure how closely AI systems follow defined values, safety policies, and governance requirements, supporting risk management, regulatory compliance, and model selection across development and deployment workflows.

• AI Alignment Framework

  AI alignment framework is a structured approach that connects organizational goals and constraints to the design, training, and governance of AI systems, enabling enterprises to specify desired behaviors, enforce safeguards, and manage risk within existing AI, data, and compliance architectures.

• AI application cybersecurity

  AI application cybersecurity is the discipline that protects enterprise AI models, data, and pipelines from security threats across development, training, deployment, and operation, enabling organizations to run AI-enabled applications while managing technical, operational, and regulatory risks around those systems.