Glossary

   –––––  THE GLOSSARY · AN INDEPENDENT PUBLICATION ON AI & LAW

Legal Tech & AI,
explained in plain words.

From the AI Act to retrieval-augmented generation:

clear, practice-oriented definitions of the terms shaping tomorrow’s legal world.

TERM OF THE WEEK

AI EXPLAIED

Hallucination

When a language model produces output that sounds plausible but is factually wrong – such as invented case law, non-existent statutes or false citations.

Why it matters: every AI-generated legal statement needs to be verified against primary sources before it is relied on.

Topic

A

B

C

D

E

F

G

H

I

J

K

L

M

N

O

P

Q

R

S

T

U

V

W

X

Y

Z

A

Algorithm

An algorithm is a set of instructions or rules designed to solve a problem or perform a specific task. Algorithms form the foundation of computer software and are used in artificial intelligence to process information, recognise patterns and support automated decisions.

Algorithmic Bias

Algorithmic bias refers to systematic distortions in the outputs or decisions of an algorithm that may lead to unfair or discriminatory outcomes. Such bias can arise from training data, model design or the way a system is used.

Algorithmic Decision-Making

Algorithmic decision-making involves using computational systems to make or support decisions based on data and predefined rules or learned patterns. It is increasingly relevant in areas such as recruitment, financial services and public administration, raising questions about transparency, accountability and fundamental rights.

AI Agent

An AI agent is a system designed to pursue specific goals by performing tasks, making decisions and interacting with its environment. Unlike a conventional chatbot, an AI agent may use external tools and execute multiple steps with varying degrees of autonomy.

AI Alignment

AI alignment refers to efforts to ensure that artificial intelligence systems behave in accordance with intended goals, instructions and relevant human values. It is a central challenge in AI safety, particularly as systems become more capable and autonomous.

AI Explainability

AI explainability refers to methods that help people understand how an artificial intelligence system produces particular outputs or decisions. It is especially relevant when AI systems are used in legally sensitive areas such as employment, healthcare or financial services.

AI Governance

AI governance encompasses the policies, processes and organisational structures used to oversee the development and use of artificial intelligence. It aims to manage risks, establish accountability and support compliance with applicable laws and ethical standards.

AI Hallucination

An AI hallucination occurs when a generative AI system produces information that is false, misleading or unsupported but presents it as a plausible response. In legal practice, hallucinations can be particularly problematic when AI tools invent court decisions, legal provisions or references.

AI Literacy

AI literacy refers to the knowledge and skills needed to understand, use and critically assess artificial intelligence systems. Under Article 4 of the EU AI Act, providers and deployers must take measures to ensure a sufficient level of AI literacy among relevant personnel and other persons operating AI systems on their behalf, taking account of the circumstances specified in that provision.

AI Model

An AI model is a computational model that has been trained or otherwise developed to identify patterns, generate outputs or perform particular tasks. Models can be integrated into broader AI systems, such as chatbots, image generators or predictive analytics tools.

AI Provider

Under the EU AI Act, a provider is a natural or legal person, public authority, agency or other body that develops an AI system or general-purpose AI model, or has one developed, and places it on the market or puts the AI system into service under its own name or trademark. The provider’s regulatory obligations depend on the type of AI system or model and the applicable provisions of the Act.

AI Risk Management

AI risk management is the process of identifying, assessing, mitigating and monitoring risks associated with artificial intelligence systems. Under the EU AI Act, providers of high-risk AI systems are generally required to establish a risk management system in accordance with Article 9.

B

Backpropagation

Backpropagation is a method used to train artificial neural networks by calculating how much each model parameter contributes to an error. This information helps the learning algorithm adjust the network’s parameters to improve its predictions.

Benchmarking

AI benchmarking is the process of evaluating an artificial intelligence model or system against standardised tests, datasets or performance criteria. Benchmarks can measure capabilities such as reasoning, accuracy and coding, although strong benchmark results do not necessarily guarantee reliable performance in real-world applications.

Bias in AI

Bias in AI refers to systematic distortions in an artificial intelligence system that may affect the accuracy, fairness or reliability of its outputs. Bias can arise from unrepresentative training data, model design or deployment conditions and may reinforce existing inequalities.

Big Data

Big Data refers to datasets characterised by a scale, complexity or rate of generation that can make them difficult to process using conventional data management techniques. Large datasets can support AI development by providing information for training models, identifying patterns and generating predictions.

Biometric Categorisation

Biometric categorisation involves assigning individuals to categories based on their biometric data, such as facial or other physical characteristics. Under the EU AI Act, certain uses of biometric categorisation are prohibited, particularly where sensitive characteristics are inferred under the conditions specified in Article 5.

Biometric Data

Biometric data is personal data resulting from specific technical processing of a person’s physical, physiological or behavioural characteristics that allows or confirms their unique identification. Examples include facial recognition templates and fingerprints, which may be subject to heightened protection under the GDPR.

Biometric Identification

Biometric identification uses biometric information to establish a person’s identity by comparing their characteristics against stored biometric data. Technologies such as facial recognition can enable identification, raising important questions about privacy, surveillance and fundamental rights.

Black Box AI

Black Box AI describes artificial intelligence systems whose internal decision-making processes are difficult for humans to understand or interpret. This lack of transparency can create challenges when AI-generated outcomes need to be explained, audited or legally challenged.

Blockchain

Blockchain is a type of distributed ledger technology that records transactions or other information in cryptographically linked blocks. Although commonly associated with cryptocurrencies, blockchain can also support applications involving digital identity, record verification and smart contracts.

Business Process Automation (BPA)

Business Process Automation involves using technology to automate repetitive or structured business activities. In legal practice, it can streamline administrative workflows, document handling and client onboarding, reducing manual work and improving consistency.

Bayesian Network

A Bayesian network is a probabilistic model that represents relationships between variables using a graphical structure. It allows systems to estimate probabilities and update predictions when new information becomes available, making it useful for reasoning under uncertainty.

Biometric Verification

Biometric verification is the process of confirming a person’s claimed identity by comparing their biometric characteristics with a stored reference. Unlike biometric identification, which determines who a person is, verification typically checks whether someone is the person they claim to be.

C

Add Your Heading Text Here

Lorem ipsum dolor sit amet, consectetur adipiscing elit. Ut elit tellus, luctus nec ullamcorper mattis, pulvinar dapibus leo.

D

Add Your Heading Text Here

Lorem ipsum dolor sit amet, consectetur adipiscing elit. Ut elit tellus, luctus nec ullamcorper mattis, pulvinar dapibus leo.

E

Add Your Heading Text Here

Lorem ipsum dolor sit amet, consectetur adipiscing elit. Ut elit tellus, luctus nec ullamcorper mattis, pulvinar dapibus leo.

F

Add Your Heading Text Here

Lorem ipsum dolor sit amet, consectetur adipiscing elit. Ut elit tellus, luctus nec ullamcorper mattis, pulvinar dapibus leo.

G

Add Your Heading Text Here

Lorem ipsum dolor sit amet, consectetur adipiscing elit. Ut elit tellus, luctus nec ullamcorper mattis, pulvinar dapibus leo.

H

Add Your Heading Text Here

Lorem ipsum dolor sit amet, consectetur adipiscing elit. Ut elit tellus, luctus nec ullamcorper mattis, pulvinar dapibus leo.

I

Add Your Heading Text Here

Lorem ipsum dolor sit amet, consectetur adipiscing elit. Ut elit tellus, luctus nec ullamcorper mattis, pulvinar dapibus leo.

J

Add Your Heading Text Here

Lorem ipsum dolor sit amet, consectetur adipiscing elit. Ut elit tellus, luctus nec ullamcorper mattis, pulvinar dapibus leo.

K

Add Your Heading Text Here

Lorem ipsum dolor sit amet, consectetur adipiscing elit. Ut elit tellus, luctus nec ullamcorper mattis, pulvinar dapibus leo.

L

Add Your Heading Text Here

Lorem ipsum dolor sit amet, consectetur adipiscing elit. Ut elit tellus, luctus nec ullamcorper mattis, pulvinar dapibus leo.

M

Add Your Heading Text Here

Lorem ipsum dolor sit amet, consectetur adipiscing elit. Ut elit tellus, luctus nec ullamcorper mattis, pulvinar dapibus leo.

N

Add Your Heading Text Here

Lorem ipsum dolor sit amet, consectetur adipiscing elit. Ut elit tellus, luctus nec ullamcorper mattis, pulvinar dapibus leo.

O

Add Your Heading Text Here

Lorem ipsum dolor sit amet, consectetur adipiscing elit. Ut elit tellus, luctus nec ullamcorper mattis, pulvinar dapibus leo.

P

Add Your Heading Text Here

Lorem ipsum dolor sit amet, consectetur adipiscing elit. Ut elit tellus, luctus nec ullamcorper mattis, pulvinar dapibus leo.

Q

Add Your Heading Text Here

Lorem ipsum dolor sit amet, consectetur adipiscing elit. Ut elit tellus, luctus nec ullamcorper mattis, pulvinar dapibus leo.

R

Add Your Heading Text Here

Lorem ipsum dolor sit amet, consectetur adipiscing elit. Ut elit tellus, luctus nec ullamcorper mattis, pulvinar dapibus leo.

S

Add Your Heading Text Here

Lorem ipsum dolor sit amet, consectetur adipiscing elit. Ut elit tellus, luctus nec ullamcorper mattis, pulvinar dapibus leo.

T

Add Your Heading Text Here

Lorem ipsum dolor sit amet, consectetur adipiscing elit. Ut elit tellus, luctus nec ullamcorper mattis, pulvinar dapibus leo.

U

Add Your Heading Text Here

Lorem ipsum dolor sit amet, consectetur adipiscing elit. Ut elit tellus, luctus nec ullamcorper mattis, pulvinar dapibus leo.

V

Add Your Heading Text Here

Lorem ipsum dolor sit amet, consectetur adipiscing elit. Ut elit tellus, luctus nec ullamcorper mattis, pulvinar dapibus leo.

W

Add Your Heading Text Here

Lorem ipsum dolor sit amet, consectetur adipiscing elit. Ut elit tellus, luctus nec ullamcorper mattis, pulvinar dapibus leo.

X

Add Your Heading Text Here

Lorem ipsum dolor sit amet, consectetur adipiscing elit. Ut elit tellus, luctus nec ullamcorper mattis, pulvinar dapibus leo.

Y

Add Your Heading Text Here

Lorem ipsum dolor sit amet, consectetur adipiscing elit. Ut elit tellus, luctus nec ullamcorper mattis, pulvinar dapibus leo.

Z

Add Your Heading Text Here

Lorem ipsum dolor sit amet, consectetur adipiscing elit. Ut elit tellus, luctus nec ullamcorper mattis, pulvinar dapibus leo.