cybercrime
Cybercrime is illegal activity in which computers, networks, digital accounts, or online services are the target, tool, or environment of the offense.
AI Terminology
Search clear AI terminology definitions, acronyms, related concepts and the reviewed videos that explain them.
Showing 561–580 of 1136 terms
Clear filterscybersecurity evaluation
A cybersecurity evaluation measures how an AI system performs on security tasks and how its capabilities could create or reduce cyber risk.
cybersecurity safeguard
An AI cybersecurity safeguard is a control that protects AI models, data, tools, infrastructure, and users from unauthorized access, manipulation, or abuse.
data augmentation
Data augmentation creates useful variations of training examples so a model can learn to handle conditions that are scarce or unevenly represented in the original data.
data breach
A data breach is an incident in which protected information is accessed, disclosed, altered, stolen, or lost without authorization.
data center cooling
Data center cooling removes heat from computing equipment so AI hardware can operate reliably within safe temperature limits.
data center moratorium
A data center moratorium is a temporary prohibition or pause on approving or constructing data center projects in a defined area.
data center noise pollution
Data center noise pollution is unwanted sound from facility cooling, electrical and backup systems that affects nearby people or environments.
data curation
Data curation is the deliberate process of selecting, organizing, cleaning, documenting and maintaining data for AI development and evaluation.
data debugging
Data debugging finds and corrects problems in the data used to train, evaluate, or operate AI systems.
data diode
A data diode is a hardware-enforced one-way communication device that permits data to flow in only one direction between security zones.
data enrichment
Data enrichment adds useful attributes from additional sources to make an existing dataset more complete and actionable.
data flow
Data flow describes how information moves through a program, including where it enters, changes, is stored, and produces an output.
data minimization
Data minimization limits information collection, use, access, and retention to what is necessary for a defined purpose.
data ownership
Data ownership defines who has authority and responsibility over stored information, including its access, portability, retention, and disposal.
data partitioning
Data partitioning divides a dataset or tensor into defined pieces so different processors, workers, or storage nodes can handle them independently or in parallel.
data quarantine
Data quarantine isolates incoming records from trusted processing until validation confirms that they meet security, format and provenance requirements.
data redaction
Data redaction removes or irreversibly masks sensitive portions of a record before the remaining information is stored, shared or processed.
data residency
AI data residency defines the geographic or infrastructure location where AI inputs, outputs and related records are processed or stored.
data synchronization
Data synchronization keeps related information consistent across systems by propagating and reconciling changes according to defined rules.