output curation
AI output curation
AI output curation is the process of reviewing, comparing and selecting model-generated material for a defined purpose.
AI Terminology
Search clear AI terminology definitions, acronyms, related concepts and the reviewed videos that explain them.
Showing 541–560 of 705 terms
Clear filtersoutput homogenization
AI output homogenization is the tendency for AI-assisted work to converge on similar patterns, language or solutions across different users and organizations.
output quality assurance
AI output QAAI output quality assurance uses checks and review to verify that AI-generated work meets defined requirements.
package-embedded agent documentation
package-local AI documentationPackage-embedded agent documentation is authoritative guidance shipped inside a software package so coding agents can read instructions that match the installed version without depending on a website.
partial artificial intelligence automation
human-AI partial automationpartial AI automationPartial AI automation delegates selected parts of a process to AI while people continue performing the remaining work.
peer-to-peer artificial intelligence network
A peer-to-peer AI network connects independently operated machines so they can contribute or consume model capacity without all computation residing in one data center.
Perplexity Computer
Perplexity Computer is an agent-oriented product that coordinates models and tools to research information, connect applications and produce multi-step deliverables.
personalized artificial intelligence model routing
Personalized AI model routing adapts model selection to an individual user's preferences, history or requirements.
personally identifiable information
PIIPersonally identifiable information is data that directly identifies a person or can reasonably be combined with other data to identify them.
physical automation
Physical automation uses machines and control systems to perform actions in the physical world with reduced direct human operation.
platform dependency risk
AI platform dependency riskAI platform riskAI platform dependency risk is the operational or commercial exposure created when a workflow relies heavily on one model vendor or serving platform.
policy analysis
AI policy analysisAI policy analysis uses AI to examine proposed or existing public policies, evidence and likely effects while leaving accountable decisions to people and institutions.
policy engine
An AI policy engine evaluates AI activity against rules and authorizes defined controls or interventions.
post-hoc artificial intelligence citation generation
post-hoc AI citation generationPost-hoc AI citation generation drafts a conclusion first and searches afterward for pages that appear to support it.
prediction market
A prediction market lets participants trade contracts whose outcomes depend on whether a specified future event occurs.
prefill-decode disaggregation
P-D disaggregationPrefill-decode disaggregation runs prompt processing and token generation on separate worker pools so each stage can be scaled and optimized independently.
primary source verification
Primary source verification checks an original document, dataset, statement, or observation to confirm that it supports the claim attributed to it.
principle of least privilege
PoLPThe principle of least privilege gives a user, process or agent only the access needed for its current authorized task.
prior probability
A prior probability represents what is believed about an uncertain quantity or event before the current evidence is incorporated.
privacy-preserving artificial intelligence inference
Privacy-preserving AI inference protects prompts, model inputs, outputs, and associated metadata from parties that are not authorized to inspect them.
Knowledge graph
Explore AI concepts
Follow general terms into specialised sub-terms. Select any node to open its definition.