model quantization
AI model quantization represents model values with lower numerical precision to reduce memory and often speed up inference.
AI Terminology
Search clear AI terminology definitions, acronyms, related concepts and the reviewed videos that explain them.
Showing 121–140 of 705 terms
Clear filtersmonitoring agent
An AI monitoring agent observes systems or information sources, interprets signals, and reports or responds to relevant changes.
multi-robot collaboration
Multi-robot collaboration is the coordinated use of two or more robots that communicate and divide work toward a shared objective.
natural language interface
NLIA natural language interface lets a person operate or configure a system using ordinary written or spoken language.
on-device inference
On-device inference runs an AI model on local hardware instead of sending every request to a remote cloud service.
real-world artificial intelligence evaluation
Real-world AI evaluation measures model or agent behavior in genuine deployment conditions where actions encounter actual users, systems, constraints, and consequences.
recursive artificial intelligence improvement
recursive AI improvementRecursive AI improvement is a feedback process in which AI helps develop better AI systems that then contribute to further improvement.
replay testing
AI replay testing reruns a recorded AI interaction or workflow to compare behavior under controlled changes.
specialized artificial intelligence agent
A specialized AI agent is configured for a defined role, task domain and set of tools rather than unrestricted general work.
tool allowlist
A tool allowlist is an explicit set of tools that a user, agent, role, or task is permitted to invoke, with all unlisted tools denied by default.
tool calling
AI tool callingAI tool calling lets a model request a structured operation from software instead of only returning natural-language text.
training data
AI training data is the collection of examples used to adjust a model's parameters and shape its learned behavior.
trajectory evaluation
AI trajectory evaluationAI trajectory evaluation assesses the sequence of reasoning-relevant states, tool calls, decisions, and side effects that led to an agent's final result.
acceptance criterion
An acceptance criterion is a specific observable condition that must be satisfied before work can be considered complete.
adoption friction
AI adoption frictionAI adoption friction is the effort, uncertainty, disruption, or risk that makes people delay or avoid moving an AI capability into routine use.
agent audit trail
An AI agent audit trail is a durable record of an agent's identity, inputs, decisions, tool calls, approvals and outcomes.
agent execution environment
An AI agent execution environment supplies and controls the tools, data, compute, state and permissions an agent can use.
agent instruction file
An agent instruction file is a repository document that defines durable rules, workflow constraints, conventions, and verification expectations for coding agents.
agent payment authorization
AI agent payment authorizationAgent payment authorization defines whether and under what limits a software agent may prepare or execute a payment for a user.
agent sandbox
An AI agent sandbox is an isolated execution environment that limits which files, processes, networks, credentials, and external systems an agent can access.
Knowledge graph
Explore AI concepts
Follow general terms into specialised sub-terms. Select any node to open its definition.