Glossary

Search results

AI Auditing(also: Algorithmic Auditing, AI Audit)

The systematic evaluation of an AI system's outputs, behaviour, or training data to identify harms such as bias, stereotype reproduction, or accessibility failures. Audits may be conducted by industry professionals, external researchers, regulators, or end users, and are…

ASL Linguistic Markers(also: Non-Manual Markers, Non-Manual Signals, ASL Facial Grammar)

Facial expressions, head movements, and body postures that serve grammatical and semantic functions in American Sign Language and other sign languages, distinct from emotional facial expressions. Common ASL linguistic markers include MM (meaning effortlessly or regularly,…

Ability Bias(also: Ability-Based Bias, Disability Bias)

A form of social bias encoded in artificial intelligence systems, particularly large language models, that reflects stereotypical or discriminatory assumptions about people with disabilities. Ability bias manifests through linguistic associations that link specific disabilities…

Algorithmic Hiring(also: AI Hiring, Hiring AI, AI-Enabled Hiring)

The use of algorithmic systems — including machine learning and large language models — to source, screen, rank, or select job candidates. Proponents argue algorithmic hiring reduces human bias and scales review; critics show it can amplify bias against disabled, Black, female,…

Automated Employment Decision System(also: AEDS, AEDT, Automated Employment Decision Tool)

A software system that screens, evaluates, categorises, recommends, or otherwise makes or facilitates hiring or employment decisions about job candidates or workers. AEDSs span résumé sorters, personality tests, gamified cognitive assessments, situational-judgement tests,…

Computer Says No(also: Computer-Says-No)

A pattern in which an organisation invokes an algorithmic or automated decision as justification for an adverse outcome — a rejected application, a denied claim, an adjusted score — thereby deflecting responsibility from human decision-makers onto the technical system.…

Counterfactual Prompting(also: Counterfactual Debiasing, Counterfactual Data Augmentation)

A bias mitigation technique that involves modifying prompts or training examples by swapping identity-related attributes (such as disability status, gender, or race) while keeping all other context identical, in order to expose and counteract biased associations in language…

Data Representativeness(also: Dataset Representativeness, Demographic Representativeness)

The degree to which a dataset reflects the diversity of the population it is intended to serve, particularly across demographic dimensions such as age, gender, race, ethnicity, disability, and socioeconomic status. In AI and machine learning, unrepresentative training data leads…

Data Stewardship(also: Dataset Stewardship, Data Governance)

The responsible management of data throughout its lifecycle, including decisions about collection, storage, access, sharing, and disposal. In accessibility research, participatory data stewardship involves disabled data contributors in decisions about how their data is used,…

Dataset Bias(also: Training Data Bias, Data Representation Bias, Sampling Bias)

A systematic skew in the composition of training data used to build machine learning models, resulting in models that perform well for overrepresented groups but poorly for underrepresented ones. In accessibility contexts, dataset bias is a pervasive problem: activity…

Digitized Assessment(also: Digitised Assessment, Digital Hiring Assessment, Computer-Based Employment Assessment)

A computer-based hiring test used by employers to evaluate candidates' personality, cognition, skills, or judgement. Common formats include personality inventories, gamified cognitive tasks (balloon-inflating risk tests, Flanker attention tasks, arithmetic mini-games),…

Disability Stereotyping(also: Disability Stereotype, Ableist Stereotyping)

The attribution of fixed, oversimplified characteristics to individuals based on their disability status. In the context of AI and language models, disability stereotyping occurs when systems associate specific disabilities with particular traits — for example, linking autism…

EU AI Act(also: European Union Artificial Intelligence Act, Artificial Intelligence Act (EU))

A European Union regulation, adopted in 2024, that establishes a risk-based framework for AI systems deployed in the EU. High-risk systems — including AI used in employment, hiring, worker management, education, and access to essential services — are subject to obligations…

Emotional Intelligence Test(also: EI Test, I-EQ Test, Emotional Intelligence Assessment)

A hiring assessment that asks candidates to identify emotions in photographs of faces or to judge appropriate emotional responses in social scenarios. Emotional-intelligence tests are particularly inaccessible to blind and low-vision candidates, who may be unable to interpret…

End-User Auditing(also: User-Led Auditing, End User Audits)

An approach to AI auditing in which everyday users — rather than professional evaluators — identify problems, biases, or harms in AI outputs based on their lived experience. End-user auditing is particularly valuable for surfacing harms against minoritised communities (including…

Facial Expression Analysis(also: Automated Facial Expression Analysis, Facial Coding, AFEA)

The automated classification of a person's facial movements into discrete emotion categories (happy, angry, neutral, surprised, etc.) using computer vision. In hiring, facial expression analysis is embedded in AI-scored video interviews. It has been shown to systematically…

IncluSet

A dataset surfacing repository created by researchers at the University of Maryland that catalogs and organizes accessibility datasets — datasets sourced from people with disabilities and older adults. IncluSet was developed to make it easier for AI researchers and practitioners…

Personality Test(also: Personality Inventory, Personality Assessment)

A psychometric instrument that attempts to quantify traits such as conscientiousness, extraversion, emotional stability, or risk tolerance. Personality tests were introduced in industry during and after World War I to screen for "maladjusted" workers, and are now the most…

Self-Debiasing(also: Model Self-Debiasing, Autonomous Debiasing)

A class of techniques where AI systems, particularly large language models, are prompted or configured to identify and reduce their own biased outputs without external model modification or retraining. Self-debiasing approaches include prompting models to reflect on whether…

Situational Judgment Test(also: SJT, Situational Judgement Test)

A hiring assessment that presents candidates with hypothetical workplace scenarios and asks them to select the "best" and "worst" responses from a predefined multiple-choice list. SJTs assume a single correct behaviour per scenario, which can systematically disadvantage…

Speech Diversity(also: Diverse Speech, Non-Typical Speech)

The full range of ways human speech varies from the narrow 'typical' speech on which most speech-AI systems are trained and benchmarked. Speech diversity includes people who stutter, d/Deaf and Hard-of-Hearing speakers, people with dysarthria, aphasia, or other neurological…

21 results.