--- canonical: https://www.jkconcepts.at/en/ai-glossary/ robots: noindex, follow --- AI glossary # AI glossary: artificial intelligence explained simply What is an LLM? What does RAG mean? And how do AI agents, chatbots and classic automation differ? Terms around AI, language models, AI search and automation – explained simply and technically. [Browse the terms](https://www.jkconcepts.at/en/ai-glossary/#glossar) [All knowledge areas](https://www.jkconcepts.at/en/knowledge/) At a glance ### Scope 80 terms ### Reading levels Simple and technical ### Last updated October 2026 Look it up instead of collecting buzzwords ## AI terms and questions from A to Z Search directly for a term or a question. Every answer includes a clear definition, practical context and relevant cross-references. Search for a term or question Topic **80** terms shown Quick start ## Frequently searched AI terms Key questions about AI, language models, agents, data and automation. [Artificial Intelligence (AI)](https://www.jkconcepts.at/en/ai-glossary/#ki) [Generative AI](https://www.jkconcepts.at/en/ai-glossary/#generative-ai) [Machine Learning (ML)](https://www.jkconcepts.at/en/ai-glossary/#machine-learning) [Deep Learning](https://www.jkconcepts.at/en/ai-glossary/#deep-learning) [LLM](https://www.jkconcepts.at/en/ai-glossary/#llm) [AI Agent](https://www.jkconcepts.at/en/ai-glossary/#ai-agent) [Prompt Engineering](https://www.jkconcepts.at/en/ai-glossary/#prompt-engineering) [RAG](https://www.jkconcepts.at/en/ai-glossary/#rag) [Hallucination](https://www.jkconcepts.at/en/ai-glossary/#hallucination) [Embedding](https://www.jkconcepts.at/en/ai-glossary/#embedding) [Vector Database](https://www.jkconcepts.at/en/ai-glossary/#vector-database) [MCP](https://www.jkconcepts.at/en/ai-glossary/#mcp) [Local AI / Local LLM](https://www.jkconcepts.at/en/ai-glossary/#local-ai) For websites & businesses ### AI & visibility on the web Terms on discoverability, machine-readable content, processes and secure integration. [AI Visibility](https://www.jkconcepts.at/en/ai-glossary/#ai-visibility) [GEO](https://www.jkconcepts.at/en/ai-glossary/#geo) [AEO](https://www.jkconcepts.at/en/ai-glossary/#aeo) [AI Search & Generative Search](https://www.jkconcepts.at/en/ai-glossary/#ai-search) [Entity](https://www.jkconcepts.at/en/ai-glossary/#entity) [Structured Data](https://www.jkconcepts.at/en/ai-glossary/#structured-data) [Schema.org](https://www.jkconcepts.at/en/ai-glossary/#schema-org) [Crawler & AI Crawler](https://www.jkconcepts.at/en/ai-glossary/#ai-crawler) [Knowledge Graph](https://www.jkconcepts.at/en/ai-glossary/#knowledge-graph) [Prompt Injection](https://www.jkconcepts.at/en/ai-glossary/#prompt-injection) [Data Protection](https://www.jkconcepts.at/en/ai-glossary/#datenschutz) ## Terms starting with A AI search · Advanced [#aeo](https://www.jkconcepts.at/en/ai-glossary/#aeo) ### What is answer engine optimisation (AEO)? AEO · Answer Engine Optimisation Preparing content so that search and answer systems can answer specific questions as unambiguously as possible. ### Technology, example and context #### Example A page answers a common customer question directly, completely and with traceable evidence. #### Why does it matter? AEO overlaps with SEO and GEO; clear information architecture, precise answers and technical readability help. #### Related terms [GEO](https://www.jkconcepts.at/en/ai-glossary/#geo) [AI Search & Generative Search](https://www.jkconcepts.at/en/ai-glossary/#ai-search) [Structured Data](https://www.jkconcepts.at/en/ai-glossary/#structured-data) Agents & automation · Advanced [#agentic-ai](https://www.jkconcepts.at/en/ai-glossary/#agentic-ai) ### What does agentic AI mean? Agentic AI · Agent-based AI An approach in which AI systems plan in several steps, act, check results and adjust their next step. ### Technology, example and context #### Example A system researches data, compares options, asks for approval when uncertain and documents the result. #### Why does it matter? More autonomy increases the potential benefit and at the same time the surface for errors, the costs and the requirements for oversight and observability. #### Related terms [AI Agent](https://www.jkconcepts.at/en/ai-glossary/#ai-agent) [Orchestration](https://www.jkconcepts.at/en/ai-glossary/#orchestration) [Human in the Loop](https://www.jkconcepts.at/en/ai-glossary/#human-in-the-loop) Agents & automation · Advanced [#ai-agent](https://www.jkconcepts.at/en/ai-glossary/#ai-agent) ### What is an AI agent? AI Agent · AI that acts towards a goal An AI system that interprets a goal, selects next steps and uses tools or data sources to do so. ### Technology, example and context #### Example An agent reads an enquiry, checks an order status via an API and drafts a suitable reply. #### Why does it matter? Agents can support flexible workflows, but need limited permissions, logging and clear approval points. #### Related terms [Agentic AI](https://www.jkconcepts.at/en/ai-glossary/#agentic-ai) [Tool & Function Calling](https://www.jkconcepts.at/en/ai-glossary/#tool-calling) [AI Guardrails](https://www.jkconcepts.at/en/ai-glossary/#ai-guardrails) Agents & automation · Advanced [#ai-automation](https://www.jkconcepts.at/en/ai-glossary/#ai-automation) ### What does AI Automation mean? AI Automation · AI-assisted automation Automation in which AI processes unstructured content or prepares decisions. ### Technology, example and context #### Example Incoming emails are summarised and categorised before a fixed workflow forwards them. #### Why does it matter? AI extends automation to less clear-cut tasks, but needs thresholds, exception paths and human oversight. #### Related terms [Workflow Automation](https://www.jkconcepts.at/en/ai-glossary/#workflow-automation) [Human in the Loop](https://www.jkconcepts.at/en/ai-glossary/#human-in-the-loop) [Structured Output](https://www.jkconcepts.at/en/ai-glossary/#structured-output) Security · Advanced [#ai-guardrails](https://www.jkconcepts.at/en/ai-glossary/#ai-guardrails) ### What does AI Guardrails mean? AI Guardrails · Technical and organisational guardrails Rules, checks and limits around the inputs, outputs and actions of an AI system. ### Technology, example and context #### Example An application blocks sensitive data, validates outputs and requires approval before external actions. #### Why does it matter? Guardrails reduce certain risks, but must be implemented in several layers and tested regularly against real failure cases. #### Related terms [Human in the Loop](https://www.jkconcepts.at/en/ai-glossary/#human-in-the-loop) [Prompt Injection](https://www.jkconcepts.at/en/ai-glossary/#prompt-injection) [Evaluation, Evals & Benchmarks](https://www.jkconcepts.at/en/ai-glossary/#evaluation-evals) AI search · Advanced [#ai-overview](https://www.jkconcepts.at/en/ai-glossary/#ai-overview) ### What does AI Overview mean? AI Overview · AI-generated search summary A summary generated by a search engine that appears above or within classic search results. ### Technology, example and context #### Example A search engine answers a question directly and links selected sources next to the summary. #### Why does it matter? AI Overviews change click paths and visibility; clear sources, reliable content and measurable mentions become more important. #### Related terms [AI Search & Generative Search](https://www.jkconcepts.at/en/ai-glossary/#ai-search) [AI Visibility](https://www.jkconcepts.at/en/ai-glossary/#ai-visibility) [Citation](https://www.jkconcepts.at/en/ai-glossary/#citation) AI search · Advanced [#ai-search](https://www.jkconcepts.at/en/ai-glossary/#ai-search) ### What is AI search? AI Search & Generative Search · AI-assisted and generative search Search that uses AI to understand information, combine it or formulate it as a direct answer. ### Technology, example and context #### Example Instead of only showing links, a search interface combines several sources into one answer. #### Why does it matter? Businesses need to provide content so clearly and accessibly that search systems can find, understand and attribute it correctly. #### Related terms [AI Overview](https://www.jkconcepts.at/en/ai-glossary/#ai-overview) [GEO](https://www.jkconcepts.at/en/ai-glossary/#geo) [Citation](https://www.jkconcepts.at/en/ai-glossary/#citation) AI search · Advanced [#ai-visibility](https://www.jkconcepts.at/en/ai-glossary/#ai-visibility) ### What does AI visibility mean? AI Visibility · Visibility in AI answers A measure of whether and how a brand, organisation or website appears in AI-generated answers. ### Technology, example and context #### Example For relevant questions, a company is named correctly and its website is cited as a source. #### Why does it matter? AI visibility covers mention, context and source reference; it should be measured with documented test questions rather than individual examples. #### Related terms [GEO](https://www.jkconcepts.at/en/ai-glossary/#geo) [Citation](https://www.jkconcepts.at/en/ai-glossary/#citation) [Entity](https://www.jkconcepts.at/en/ai-glossary/#entity) Development · Technical [#api](https://www.jkconcepts.at/en/ai-glossary/#api) ### What is an API? API · Application Programming Interface A defined interface through which software systems exchange data or call functions. ### Technology, example and context #### Example A website sends text to an AI model and receives the answer back via its API. #### Why does it matter? APIs enable controlled integration into existing systems; costs, limits, availability and data transfer must be planned for. #### Related terms [Tool & Function Calling](https://www.jkconcepts.at/en/ai-glossary/#tool-calling) [MCP](https://www.jkconcepts.at/en/ai-glossary/#mcp) [Cloud AI](https://www.jkconcepts.at/en/ai-glossary/#cloud-ai) Basics · Basic [#ki](https://www.jkconcepts.at/en/ai-glossary/#ki) ### What is artificial intelligence (AI)? Artificial Intelligence (AI) · Machines performing cognitive tasks Systems that perform tasks that would normally require human abilities such as understanding, learning or deciding. ### Technology, example and context #### Example An application sorts support requests, recognises their topic and suggests suitable replies. #### Why does it matter? AI is an umbrella term. For a sensible assessment, it always has to be clarified which concrete task, data basis and oversight are meant. #### Related terms [Machine Learning (ML)](https://www.jkconcepts.at/en/ai-glossary/#machine-learning) [Generative AI](https://www.jkconcepts.at/en/ai-glossary/#generative-ai) [AI Automation](https://www.jkconcepts.at/en/ai-glossary/#ai-automation) ## Terms starting with B Security · Advanced [#bias](https://www.jkconcepts.at/en/ai-glossary/#bias) ### What does Bias mean? Bias · Systematic distortion A systematic skew in data or results that can unfairly disadvantage certain groups, cases or perspectives. ### Technology, example and context #### Example A system rates certain applicant profiles lower because historical data reflects earlier inequalities. #### Why does it matter? Bias cannot be solved technically alone; data, the definition of the goal, test groups and impacts have to be considered together. #### Related terms [Training Data](https://www.jkconcepts.at/en/ai-glossary/#training-data) [Evaluation, Evals & Benchmarks](https://www.jkconcepts.at/en/ai-glossary/#evaluation-evals) [Human in the Loop](https://www.jkconcepts.at/en/ai-glossary/#human-in-the-loop) ## Terms starting with C AI search · Advanced [#citation](https://www.jkconcepts.at/en/ai-glossary/#citation) ### What does Citation mean? Citation · Source reference A reference by an AI or search system to the source of a statement. ### Technology, example and context #### Example A generated answer links the specific expert article an item of information comes from. #### Why does it matter? Citations increase traceability and can bring qualified traffic, but don’t automatically prove a correct summary. #### Related terms [Grounding](https://www.jkconcepts.at/en/ai-glossary/#grounding) [AI Visibility](https://www.jkconcepts.at/en/ai-glossary/#ai-visibility) [AI Search & Generative Search](https://www.jkconcepts.at/en/ai-glossary/#ai-search) Development · Technical [#cloud-ai](https://www.jkconcepts.at/en/ai-glossary/#cloud-ai) ### What does Cloud AI mean? Cloud AI · Externally operated AI service An AI model operated by a provider and usually made available via a web interface or API. ### Technology, example and context #### Example An application sends approved content to a model service and receives the processing as an API response. #### Why does it matter? Cloud AI is quick to use and scalable; contracts, data flows, region, costs and provider dependency need to be assessed. #### Related terms [API](https://www.jkconcepts.at/en/ai-glossary/#api) [Local AI / Local LLM](https://www.jkconcepts.at/en/ai-glossary/#local-ai) [Data Protection](https://www.jkconcepts.at/en/ai-glossary/#datenschutz) Basics · Basic [#computer-vision](https://www.jkconcepts.at/en/ai-glossary/#computer-vision) ### What does Computer Vision mean? Computer Vision · Machine understanding of images AI methods that let computers recognise content and patterns in images or videos. ### Technology, example and context #### Example A quality check highlights visible damage on product photos. #### Why does it matter? Computer vision can support visual inspections, but must be tested with realistic lighting, camera and edge cases. #### Related terms [Multimodal AI](https://www.jkconcepts.at/en/ai-glossary/#multimodal-ai) [OCR](https://www.jkconcepts.at/en/ai-glossary/#ocr) [Deep Learning](https://www.jkconcepts.at/en/ai-glossary/#deep-learning) Language models · Advanced [#context-window](https://www.jkconcepts.at/en/ai-glossary/#context-window) ### What is a context window? Context Window · How much a model can consider at once The maximum amount of information a model can take into account in a single run. ### Technology, example and context #### Example With a long contract, relevant sections have to be selected if the whole document doesn’t fit into the context window. #### Why does it matter? A large context window helps with long content, but doesn’t replace good selection, structure or checking of the information provided. #### Related terms [Token](https://www.jkconcepts.at/en/ai-glossary/#token) [Retrieval](https://www.jkconcepts.at/en/ai-glossary/#retrieval) [RAG](https://www.jkconcepts.at/en/ai-glossary/#rag) AI search · Advanced [#ai-crawler](https://www.jkconcepts.at/en/ai-glossary/#ai-crawler) ### What does Crawler & AI Crawler mean? Crawler & AI Crawler · Automated retrieval of web pages Software that retrieves web pages automatically; AI crawlers do so for AI search, training or other AI services. ### Technology, example and context #### Example A bot retrieves publicly accessible pages and processes headings, texts and links. #### Why does it matter? Technical accessibility, robots rules and clear page structures influence which content a system can process at all. #### Related terms [AI Search & Generative Search](https://www.jkconcepts.at/en/ai-glossary/#ai-search) [GEO](https://www.jkconcepts.at/en/ai-glossary/#geo) [Structured Data](https://www.jkconcepts.at/en/ai-glossary/#structured-data) ## Terms starting with D Security · Advanced [#datenschutz](https://www.jkconcepts.at/en/ai-glossary/#datenschutz) ### What does Data Protection mean? Data Protection · Protection of personal data Rules and measures for the lawful, purpose-bound and secure processing of personal data. ### Technology, example and context #### Example Before using an AI service, the purpose, necessary data, provider, storage location, contracts and deletion periods are checked. #### Why does it matter? AI doesn’t change the principles of data protection. What matters is the specific data flow, not the label of the tool used. #### Related terms [PII](https://www.jkconcepts.at/en/ai-glossary/#pii) [Pseudonymisation](https://www.jkconcepts.at/en/ai-glossary/#pseudonymisierung) [Cloud AI](https://www.jkconcepts.at/en/ai-glossary/#cloud-ai) Basics · Basic [#deep-learning](https://www.jkconcepts.at/en/ai-glossary/#deep-learning) ### What is deep learning? Deep Learning · Learning with multi-layer neural networks Machine learning with multi-layer neural networks that can process particularly complex patterns. ### Technology, example and context #### Example Deep learning models recognise objects in images or generate natural-sounding speech. #### Why does it matter? The method enables many modern AI applications, but usually requires a lot of computing power, data and careful testing. #### Related terms [Machine Learning (ML)](https://www.jkconcepts.at/en/ai-glossary/#machine-learning) [Neural Network](https://www.jkconcepts.at/en/ai-glossary/#neural-network) [GPU & VRAM](https://www.jkconcepts.at/en/ai-glossary/#gpu-vram) Basics · Basic [#diffusion-model](https://www.jkconcepts.at/en/ai-glossary/#diffusion-model) ### What does Diffusion Model mean? Diffusion Model · Generative image model A generative model that often creates images by gradually developing a suitable output from noise. ### Technology, example and context #### Example An illustration for an early campaign concept is created from a text description. #### Why does it matter? Diffusion models are powerful for visual drafts; rights, brand references and factual image content must be checked. #### Related terms [Generative AI](https://www.jkconcepts.at/en/ai-glossary/#generative-ai) [Training Data](https://www.jkconcepts.at/en/ai-glossary/#training-data) [Synthetic Data](https://www.jkconcepts.at/en/ai-glossary/#synthetic-data) ## Terms starting with E Development · Technical [#edge-ai](https://www.jkconcepts.at/en/ai-glossary/#edge-ai) ### What does Edge AI mean? Edge AI · AI close to the device AI processing directly on or close to a device instead of sending every input to a central cloud service. ### Technology, example and context #### Example A camera detects an event locally and only transmits the result instead of the entire video stream. #### Why does it matter? Edge AI can reduce latency and data transfer, but is limited by the computing power, energy and maintainability of the devices. #### Related terms [Local AI / Local LLM](https://www.jkconcepts.at/en/ai-glossary/#local-ai) [Latency](https://www.jkconcepts.at/en/ai-glossary/#latency) [Quantisation](https://www.jkconcepts.at/en/ai-glossary/#quantization) Data · Advanced [#embedding](https://www.jkconcepts.at/en/ai-glossary/#embedding) ### What are embeddings? Embedding · Numerical representation of meaning A sequence of numbers that represents the meaning or properties of content in a way that allows similarities to be calculated. ### Technology, example and context #### Example A question and document sections are compared as embeddings to find passages that match in content. #### Why does it matter? Embeddings enable semantic search, but need a model suited to the content and protection of sensitive source data. #### Related terms [Vector](https://www.jkconcepts.at/en/ai-glossary/#vector) [Semantic Search](https://www.jkconcepts.at/en/ai-glossary/#semantic-search) [Vector Database](https://www.jkconcepts.at/en/ai-glossary/#vector-database) AI search · Advanced [#entity](https://www.jkconcepts.at/en/ai-glossary/#entity) ### What does Entity mean? Entity · A uniquely identifiable thing or concept A uniquely identifiable person, organisation, product, place or term with its own properties. ### Technology, example and context #### Example JK Concepts is an entity as a company with a founder, location, services and website. #### Why does it matter? Unique names, consistent details and relationships help search and AI systems assign content to the right entity. #### Related terms [Knowledge Graph](https://www.jkconcepts.at/en/ai-glossary/#knowledge-graph) [Structured Data](https://www.jkconcepts.at/en/ai-glossary/#structured-data) [AI Visibility](https://www.jkconcepts.at/en/ai-glossary/#ai-visibility) Security · Advanced [#evaluation-evals](https://www.jkconcepts.at/en/ai-glossary/#evaluation-evals) ### What does Evaluation, Evals & Benchmarks mean? Evaluation, Evals & Benchmarks · Systematic performance testing Planned tests that assess models or entire AI systems against defined cases and criteria. ### Technology, example and context #### Example A hundred typical and difficult requests are checked for correctness, source reference, cost and response time. #### Why does it matter? General benchmarks help with comparisons; what is decisive in addition are your own evals with real business and risk cases. #### Related terms [Model Drift](https://www.jkconcepts.at/en/ai-glossary/#model-drift) [Bias](https://www.jkconcepts.at/en/ai-glossary/#bias) [AI Guardrails](https://www.jkconcepts.at/en/ai-glossary/#ai-guardrails) ## Terms starting with F Language models · Advanced [#few-shot](https://www.jkconcepts.at/en/ai-glossary/#few-shot) ### What does Few-Shot mean? Few-Shot · Task with a few examples A model receives a few examples from which it should recognise the format or desired behaviour. ### Technology, example and context #### Example The prompt shows three correctly categorised emails before a fourth is classified. #### Why does it matter? A few good examples can increase consistency, but use up context space and must cover typical special cases. #### Related terms [Zero-Shot](https://www.jkconcepts.at/en/ai-glossary/#zero-shot) [Prompt Engineering](https://www.jkconcepts.at/en/ai-glossary/#prompt-engineering) [Context Window](https://www.jkconcepts.at/en/ai-glossary/#context-window) Language models · Advanced [#fine-tuning](https://www.jkconcepts.at/en/ai-glossary/#fine-tuning) ### What is fine-tuning? Fine-Tuning · Targeted model adaptation Additional training of an existing model for a particular style, domain or type of task. ### Technology, example and context #### Example A model is tuned with reviewed support dialogues to the desired classification and tone. #### Why does it matter? Fine-tuning can improve recurring behaviour, but is usually less suitable for current knowledge than connected data access. #### Related terms [LoRA](https://www.jkconcepts.at/en/ai-glossary/#lora) [Training](https://www.jkconcepts.at/en/ai-glossary/#training) [RAG](https://www.jkconcepts.at/en/ai-glossary/#rag) Basics · Basic [#foundation-model](https://www.jkconcepts.at/en/ai-glossary/#foundation-model) ### What does Foundation Model mean? Foundation Model · General-purpose base model A large, generally trained model that serves as the basis for many different applications. ### Technology, example and context #### Example One base model is used for text analysis, support drafts and an internal search in different applications. #### Why does it matter? Businesses usually use an existing foundation model and extend it with prompts, data access or targeted adaptation. #### Related terms [Model](https://www.jkconcepts.at/en/ai-glossary/#model) [Fine-Tuning](https://www.jkconcepts.at/en/ai-glossary/#fine-tuning) [RAG](https://www.jkconcepts.at/en/ai-glossary/#rag) ## Terms starting with G Basics · Basic [#generative-ai](https://www.jkconcepts.at/en/ai-glossary/#generative-ai) ### What is generative AI? Generative AI · AI that creates new content AI that generates new content such as text, images, audio, video or code. ### Technology, example and context #### Example A system turns bullet points into a draft product description. #### Why does it matter? Generative AI can speed up knowledge work, but it does not produce automatically verified facts and needs clear quality controls. #### Related terms [Foundation Model](https://www.jkconcepts.at/en/ai-glossary/#foundation-model) [LLM](https://www.jkconcepts.at/en/ai-glossary/#llm) [Diffusion Model](https://www.jkconcepts.at/en/ai-glossary/#diffusion-model) AI search · Advanced [#geo](https://www.jkconcepts.at/en/ai-glossary/#geo) ### What is generative engine optimisation (GEO)? GEO · Generative Engine Optimisation Optimising content and technical accessibility for generative search and answer systems. ### Technology, example and context #### Example A specialist page defines terms unambiguously, backs up statements, uses structured data and is technically easy to retrieve. #### Why does it matter? GEO aims to improve the chance of being found and used as a source correctly, but cannot guarantee a mention or placement. #### Related terms [AI Visibility](https://www.jkconcepts.at/en/ai-glossary/#ai-visibility) [Structured Data](https://www.jkconcepts.at/en/ai-glossary/#structured-data) [Entity](https://www.jkconcepts.at/en/ai-glossary/#entity) Development · Technical [#gpu-vram](https://www.jkconcepts.at/en/ai-glossary/#gpu-vram) ### What does GPU & VRAM mean? GPU & VRAM · Graphics processor and graphics memory A GPU is a highly parallel processor; VRAM is its memory for models, data and intermediate values. ### Technology, example and context #### Example A larger local model needs enough VRAM so that its weights can be loaded fully or efficiently. #### Why does it matter? GPU performance and VRAM determine which models run locally, how fast they respond and what hardware costs arise. #### Related terms [Local AI / Local LLM](https://www.jkconcepts.at/en/ai-glossary/#local-ai) [Quantisation](https://www.jkconcepts.at/en/ai-glossary/#quantization) [Inference](https://www.jkconcepts.at/en/ai-glossary/#inference) Language models · Advanced [#grounding](https://www.jkconcepts.at/en/ai-glossary/#grounding) ### What does Grounding mean? Grounding · Anchoring in reliable information Linking an AI answer to provided, verifiable information. ### Technology, example and context #### Example An assistant answers a product question only on the basis of the current documentation and names the sections used. #### Why does it matter? Grounding can make answers more traceable, but does not fully prevent errors from unsuitable sources or misinterpretation. #### Related terms [RAG](https://www.jkconcepts.at/en/ai-glossary/#rag) [Citation](https://www.jkconcepts.at/en/ai-glossary/#citation) [Hallucination](https://www.jkconcepts.at/en/ai-glossary/#hallucination) ## Terms starting with H Security · Advanced [#hallucination](https://www.jkconcepts.at/en/ai-glossary/#hallucination) ### What are AI hallucinations? Hallucination · A plausible-sounding false statement Factually wrong or invented model output that can nevertheless be phrased convincingly. ### Technology, example and context #### Example A model cites a source that doesn’t exist or invents a product feature. #### Why does it matter? Hallucinations make sources, validation and human oversight indispensable, especially for consequential statements. #### Related terms [Grounding](https://www.jkconcepts.at/en/ai-glossary/#grounding) [RAG](https://www.jkconcepts.at/en/ai-glossary/#rag) [Human in the Loop](https://www.jkconcepts.at/en/ai-glossary/#human-in-the-loop) Security · Advanced [#human-in-the-loop](https://www.jkconcepts.at/en/ai-glossary/#human-in-the-loop) ### What does Human in the Loop mean? Human in the Loop · Human oversight in the process A process design in which people check results, approve decisions or take over difficult cases. ### Technology, example and context #### Example The AI prepares a contract summary and a qualified person checks it before it is used further. #### Why does it matter? Human oversight is especially important where errors have serious consequences, but must be designed with clear information and a real ability to intervene. #### Related terms [AI Guardrails](https://www.jkconcepts.at/en/ai-glossary/#ai-guardrails) [AI Automation](https://www.jkconcepts.at/en/ai-glossary/#ai-automation) [Hallucination](https://www.jkconcepts.at/en/ai-glossary/#hallucination) ## Terms starting with I Language models · Advanced [#inference](https://www.jkconcepts.at/en/ai-glossary/#inference) ### What does inference mean in AI? Inference · Running a trained model Using a trained model to compute an output for a new input. ### Technology, example and context #### Example Every generated chat reply is an inference in which the model computes new tokens from the prompt and context. #### Why does it matter? For live applications, inference costs, speed and reliability make up a large part of operation. #### Related terms [Model](https://www.jkconcepts.at/en/ai-glossary/#model) [Training](https://www.jkconcepts.at/en/ai-glossary/#training) [Latency](https://www.jkconcepts.at/en/ai-glossary/#latency) ## Terms starting with J Security · Advanced [#jailbreak](https://www.jkconcepts.at/en/ai-glossary/#jailbreak) ### What does Jailbreak mean? Jailbreak · Attempt to bypass model safeguards An attempt to make a model ignore its safety rules through specially crafted inputs. ### Technology, example and context #### Example An input disguises a prohibited request as a role-play or splits it into several seemingly harmless steps. #### Why does it matter? Models can never be secured against new bypass attempts by prompts alone; the application and permissions must limit the damage. #### Related terms [Prompt Injection](https://www.jkconcepts.at/en/ai-glossary/#prompt-injection) [AI Guardrails](https://www.jkconcepts.at/en/ai-glossary/#ai-guardrails) [System Prompt](https://www.jkconcepts.at/en/ai-glossary/#system-prompt) ## Terms starting with K Data · Advanced [#knowledge-base](https://www.jkconcepts.at/en/ai-glossary/#knowledge-base) ### What does Knowledge Base mean? Knowledge Base · Curated collection of information A maintained collection of information that people or systems can access in a targeted way. ### Technology, example and context #### Example Product documentation, internal guidelines and approved support articles form a knowledge base. #### Why does it matter? An AI can only answer as reliably about company knowledge as the content, responsibilities and updating of the knowledge base are organised. #### Related terms [RAG](https://www.jkconcepts.at/en/ai-glossary/#rag) [Knowledge Graph](https://www.jkconcepts.at/en/ai-glossary/#knowledge-graph) [Retrieval](https://www.jkconcepts.at/en/ai-glossary/#retrieval) Data · Advanced [#knowledge-graph](https://www.jkconcepts.at/en/ai-glossary/#knowledge-graph) ### What does Knowledge Graph mean? Knowledge Graph · Knowledge as entities and relationships A representation of knowledge as uniquely named things and their relationships to each other. ### Technology, example and context #### Example A graph connects a company with its products, locations, contacts and certifications. #### Why does it matter? Knowledge graphs make relationships machine-readable and help with search, data integration and unambiguous entity assignment. #### Related terms [Entity](https://www.jkconcepts.at/en/ai-glossary/#entity) [Structured Data](https://www.jkconcepts.at/en/ai-glossary/#structured-data) [Knowledge Base](https://www.jkconcepts.at/en/ai-glossary/#knowledge-base) ## Terms starting with L Development · Technical [#latency](https://www.jkconcepts.at/en/ai-glossary/#latency) ### What does Latency mean? Latency · Response delay The time delay between a request and the start or completion of processing. ### Technology, example and context #### Example An assistant takes two seconds until the first text appears and eight seconds until the answer is complete. #### Why does it matter? High latency worsens usability and can slow down automated processes; model, network and orchestration all play a part. #### Related terms [Inference](https://www.jkconcepts.at/en/ai-glossary/#inference) [Edge AI](https://www.jkconcepts.at/en/ai-glossary/#edge-ai) [Model Distillation](https://www.jkconcepts.at/en/ai-glossary/#model-distillation) Language models · Advanced [#llm](https://www.jkconcepts.at/en/ai-glossary/#llm) ### What is a large language model (LLM)? LLM · Large Language Model A large language model that has learned patterns in text and uses them to analyse or generate content. ### Technology, example and context #### Example An LLM summarises long documents or answers questions in natural language. #### Why does it matter? LLMs are versatile, but they neither automatically know your current company data nor guarantee correct answers. #### Related terms [Transformer](https://www.jkconcepts.at/en/ai-glossary/#transformer) [Token](https://www.jkconcepts.at/en/ai-glossary/#token) [RAG](https://www.jkconcepts.at/en/ai-glossary/#rag) Development · Technical [#local-ai](https://www.jkconcepts.at/en/ai-glossary/#local-ai) ### What is local AI? Local AI / Local LLM · AI run on your own hardware An AI model that runs on your own or directly controlled hardware instead of an external model service. ### Technology, example and context #### Example A small language model processes internal texts on a company-owned server without an external API call. #### Why does it matter? Running AI locally can increase data control, but shifts responsibility for hardware, updates, security and availability to the business. #### Related terms [Cloud AI](https://www.jkconcepts.at/en/ai-glossary/#cloud-ai) [Edge AI](https://www.jkconcepts.at/en/ai-glossary/#edge-ai) [Quantisation](https://www.jkconcepts.at/en/ai-glossary/#quantization) Language models · Advanced [#lora](https://www.jkconcepts.at/en/ai-glossary/#lora) ### What does LoRA mean? LoRA · Low-Rank Adaptation A resource-efficient method for adapting a large model through a small number of additional weights. ### Technology, example and context #### Example An open model receives a small specialised adaptation without retraining all the original parameters. #### Why does it matter? LoRA reduces the effort and memory needed for an adaptation, but still requires good training data and clean evaluation. #### Related terms [Fine-Tuning](https://www.jkconcepts.at/en/ai-glossary/#fine-tuning) [Parameters](https://www.jkconcepts.at/en/ai-glossary/#parameters) [Open-Weight Model](https://www.jkconcepts.at/en/ai-glossary/#open-weight-model) ## Terms starting with M Basics · Basic [#machine-learning](https://www.jkconcepts.at/en/ai-glossary/#machine-learning) ### What is machine learning? Machine Learning (ML) · Learning patterns from data A branch of AI in which systems learn patterns from example data instead of receiving every step as a fixed rule. ### Technology, example and context #### Example A model learns from previous orders which transactions look unusual. #### Why does it matter? The quality of the results depends heavily on suitable data, testing and how similar the training cases are to real-world cases. #### Related terms [Training Data](https://www.jkconcepts.at/en/ai-glossary/#training-data) [Model](https://www.jkconcepts.at/en/ai-glossary/#model) [Deep Learning](https://www.jkconcepts.at/en/ai-glossary/#deep-learning) Agents & automation · Advanced [#mcp](https://www.jkconcepts.at/en/ai-glossary/#mcp) ### What is MCP (Model Context Protocol)? MCP · Model Context Protocol An open protocol through which AI applications can connect tools and context sources in a uniform way. ### Technology, example and context #### Example An AI assistant accesses documents or internal tools through standardised MCP interfaces. #### Why does it matter? MCP can standardise integrations, but replaces neither authentication, permission checks nor secure tool execution. #### Related terms [Tool & Function Calling](https://www.jkconcepts.at/en/ai-glossary/#tool-calling) [API](https://www.jkconcepts.at/en/ai-glossary/#api) [Orchestration](https://www.jkconcepts.at/en/ai-glossary/#orchestration) Language models · Advanced [#mixture-of-experts](https://www.jkconcepts.at/en/ai-glossary/#mixture-of-experts) ### What does Mixture of Experts (MoE) mean? Mixture of Experts (MoE) · Combination of specialised sub-models A model architecture that activates only selected specialised parts of a larger model for each input. ### Technology, example and context #### Example A routing mechanism directs different tokens to the expert layers that suit them. #### Why does it matter? MoE can make high model capacity usable more efficiently; the total parameter count does not equal the computing cost of each request. #### Related terms [Transformer](https://www.jkconcepts.at/en/ai-glossary/#transformer) [Parameters](https://www.jkconcepts.at/en/ai-glossary/#parameters) [Inference](https://www.jkconcepts.at/en/ai-glossary/#inference) Basics · Basic [#model](https://www.jkconcepts.at/en/ai-glossary/#model) ### What is an AI model? Model · AI model A trained mathematical system that produces predictions, assessments or new content from inputs. ### Technology, example and context #### Example A language model receives a question and computes an answer from it step by step. #### Why does it matter? Not every model suits every task. Quality, cost, data protection and operation differ considerably. #### Related terms [Training](https://www.jkconcepts.at/en/ai-glossary/#training) [Inference](https://www.jkconcepts.at/en/ai-glossary/#inference) [Parameters](https://www.jkconcepts.at/en/ai-glossary/#parameters) Language models · Advanced [#model-distillation](https://www.jkconcepts.at/en/ai-glossary/#model-distillation) ### What does Model Distillation mean? Model Distillation · Teaching a smaller model from a larger one A method in which a smaller model learns from the outputs or behaviour of a stronger model. ### Technology, example and context #### Example A compact model is trained for a clearly defined classification task using examples from a large model. #### Why does it matter? Distillation can reduce cost and latency, but may also transfer errors or biases of the original model. #### Related terms [Model](https://www.jkconcepts.at/en/ai-glossary/#model) [Fine-Tuning](https://www.jkconcepts.at/en/ai-glossary/#fine-tuning) [Latency](https://www.jkconcepts.at/en/ai-glossary/#latency) Security · Advanced [#model-drift](https://www.jkconcepts.at/en/ai-glossary/#model-drift) ### What does Model Drift mean? Model Drift · Declining model fit in operation A change in model quality because data, language, processes or real-world relationships differ from the time of testing. ### Technology, example and context #### Example A classification becomes less accurate after new product groups and different wording are introduced. #### Why does it matter? AI systems need ongoing samples, metrics and repeated tests; a good starting value does not guarantee lasting quality. #### Related terms [Training Data](https://www.jkconcepts.at/en/ai-glossary/#training-data) [Evaluation, Evals & Benchmarks](https://www.jkconcepts.at/en/ai-glossary/#evaluation-evals) [Bias](https://www.jkconcepts.at/en/ai-glossary/#bias) Basics · Basic [#multimodal-ai](https://www.jkconcepts.at/en/ai-glossary/#multimodal-ai) ### What is multimodal AI? Multimodal AI · AI across text, images, audio and video AI that can process several types of media such as text, images, audio or video together. ### Technology, example and context #### Example A model reads a photo of a rating plate, recognises the serial number and explains the matching maintenance instructions. #### Why does it matter? Multimodal systems open up documents and real working situations, but need clear rules for sensitive content. #### Related terms [Computer Vision](https://www.jkconcepts.at/en/ai-glossary/#computer-vision) [OCR](https://www.jkconcepts.at/en/ai-glossary/#ocr) [LLM](https://www.jkconcepts.at/en/ai-glossary/#llm) ## Terms starting with N Basics · Basic [#neural-network](https://www.jkconcepts.at/en/ai-glossary/#neural-network) ### What does Neural Network mean? Neural Network · Layered computational model A computational model of connected layers that learns weights and turns inputs into predictions or outputs. ### Technology, example and context #### Example A neural network assesses whether a document is an invoice, a quote or a delivery note. #### Why does it matter? Neural networks are the technical basis of many language, image and audio models, but remain hard to judge without evaluation. #### Related terms [Deep Learning](https://www.jkconcepts.at/en/ai-glossary/#deep-learning) [Parameters](https://www.jkconcepts.at/en/ai-glossary/#parameters) [Transformer](https://www.jkconcepts.at/en/ai-glossary/#transformer) ## Terms starting with O Basics · Basic [#ocr](https://www.jkconcepts.at/en/ai-glossary/#ocr) ### What does OCR mean? OCR · Optical Character Recognition A technique that turns printed or handwritten text from images and scans into machine-readable text. ### Technology, example and context #### Example Invoice data is recognised from a scanned PDF and prepared for accounting. #### Why does it matter? OCR makes documents searchable and automatable; recognition errors in numbers or tables still need checking. #### Related terms [Computer Vision](https://www.jkconcepts.at/en/ai-glossary/#computer-vision) [Multimodal AI](https://www.jkconcepts.at/en/ai-glossary/#multimodal-ai) [Structured Output](https://www.jkconcepts.at/en/ai-glossary/#structured-output) Development · Technical [#open-source-ai](https://www.jkconcepts.at/en/ai-glossary/#open-source-ai) ### What does Open-Source AI mean? Open-Source AI · Open AI software An AI system whose relevant components are available under open licences; the exact scope can vary greatly. ### Technology, example and context #### Example The code for training and operation is open, while the dataset or model weights may have separate conditions. #### Why does it matter? The term is used inconsistently. Before use, licences, available components and commercial rights need to be checked specifically. #### Related terms [Open-Weight Model](https://www.jkconcepts.at/en/ai-glossary/#open-weight-model) [Local AI / Local LLM](https://www.jkconcepts.at/en/ai-glossary/#local-ai) [Training Data](https://www.jkconcepts.at/en/ai-glossary/#training-data) Development · Technical [#open-weight-model](https://www.jkconcepts.at/en/ai-glossary/#open-weight-model) ### What is an open-weight model? Open-Weight Model · Model with publicly available weights A model whose trained weights can be downloaded and run on your own infrastructure. ### Technology, example and context #### Example A company runs an available language model on its own server and adjusts its configuration. #### Why does it matter? Open weights allow more operational control, but don’t automatically mean open training data, free use or low running costs. #### Related terms [Open-Source AI](https://www.jkconcepts.at/en/ai-glossary/#open-source-ai) [Local AI / Local LLM](https://www.jkconcepts.at/en/ai-glossary/#local-ai) [Parameters](https://www.jkconcepts.at/en/ai-glossary/#parameters) Agents & automation · Advanced [#orchestration](https://www.jkconcepts.at/en/ai-glossary/#orchestration) ### What does Orchestration mean? Orchestration · Coordinating models and tools Technical coordination of several models, data sources, tools and process steps. ### Technology, example and context #### Example An application chooses a model, retrieves company data, validates the output and then starts an approval step. #### Why does it matter? Good orchestration makes complex AI applications testable and operable instead of hiding logic in a single prompt. #### Related terms [AI Agent](https://www.jkconcepts.at/en/ai-glossary/#ai-agent) [Workflow Automation](https://www.jkconcepts.at/en/ai-glossary/#workflow-automation) [MCP](https://www.jkconcepts.at/en/ai-glossary/#mcp) ## Terms starting with P Language models · Advanced [#parameters](https://www.jkconcepts.at/en/ai-glossary/#parameters) ### What does Parameters mean? Parameters · Model parameters Learned numerical values in the model that determine how inputs are processed and outputs computed. ### Technology, example and context #### Example During training, a very large number of weights are adjusted so that the model makes better predictions. #### Why does it matter? More parameters can increase model capacity, but don’t automatically mean better results for a specific business case. #### Related terms [Training](https://www.jkconcepts.at/en/ai-glossary/#training) [Open-Weight Model](https://www.jkconcepts.at/en/ai-glossary/#open-weight-model) [Quantisation](https://www.jkconcepts.at/en/ai-glossary/#quantization) Security · Advanced [#pii](https://www.jkconcepts.at/en/ai-glossary/#pii) ### What does PII mean? PII · Personally Identifiable Information Information that can identify a person directly or indirectly. ### Technology, example and context #### Example Name, email address, customer number and a combination of location and contract data can be personal data. #### Why does it matter? Before AI processing, it must be clarified which personal data is really needed and where it is transferred to. #### Related terms [Data Protection](https://www.jkconcepts.at/en/ai-glossary/#datenschutz) [Pseudonymisation](https://www.jkconcepts.at/en/ai-glossary/#pseudonymisierung) [Cloud AI](https://www.jkconcepts.at/en/ai-glossary/#cloud-ai) Language models · Advanced [#prompt](https://www.jkconcepts.at/en/ai-glossary/#prompt) ### What is a prompt? Prompt · Input to an AI model An instruction, question or context that an AI model receives as input. ### Technology, example and context #### Example A prompt contains the task, the target tone, relevant facts and the desired output format. #### Why does it matter? Clear prompts make useful results more likely, but replace neither reliable data nor expert review. #### Related terms [Prompt Engineering](https://www.jkconcepts.at/en/ai-glossary/#prompt-engineering) [System Prompt](https://www.jkconcepts.at/en/ai-glossary/#system-prompt) [Structured Output](https://www.jkconcepts.at/en/ai-glossary/#structured-output) Language models · Advanced [#prompt-engineering](https://www.jkconcepts.at/en/ai-glossary/#prompt-engineering) ### What is prompt engineering? Prompt Engineering · Systematic design of prompts Planning, testing and improving inputs so that a model performs the desired task more reliably. ### Technology, example and context #### Example An instruction is extended with a role description, examples, limits and a JSON schema and then tested. #### Why does it matter? Good prompt engineering reduces misunderstandings, but should be seen as part of a tested system rather than a magic formula. #### Related terms [Prompt](https://www.jkconcepts.at/en/ai-glossary/#prompt) [Few-Shot](https://www.jkconcepts.at/en/ai-glossary/#few-shot) [Evaluation, Evals & Benchmarks](https://www.jkconcepts.at/en/ai-glossary/#evaluation-evals) Security · Advanced [#prompt-injection](https://www.jkconcepts.at/en/ai-glossary/#prompt-injection) ### What does Prompt Injection mean? Prompt Injection · Manipulation through injected instructions An attack in which malicious instructions in inputs or in content that is read are meant to manipulate the AI system. ### Technology, example and context #### Example A document contains hidden instructions intended to make an agent reveal data or call a tool. #### Why does it matter? As soon as models read external content or use tools, system prompts are not enough; permissions and actions must be limited technically. #### Related terms [Jailbreak](https://www.jkconcepts.at/en/ai-glossary/#jailbreak) [System Prompt](https://www.jkconcepts.at/en/ai-glossary/#system-prompt) [AI Guardrails](https://www.jkconcepts.at/en/ai-glossary/#ai-guardrails) Security · Advanced [#pseudonymisierung](https://www.jkconcepts.at/en/ai-glossary/#pseudonymisierung) ### What does Pseudonymisation mean? Pseudonymisation · Separating data from direct identity Processing in which direct identifiers are replaced and separate additional information is kept protected. ### Technology, example and context #### Example Before analysis, a customer number is replaced by an internal pseudonym; the mapping is kept separately protected. #### Why does it matter? Pseudonymisation can reduce risks, but doesn’t automatically make data anonymous and doesn’t remove data protection obligations. #### Related terms [Data Protection](https://www.jkconcepts.at/en/ai-glossary/#datenschutz) [PII](https://www.jkconcepts.at/en/ai-glossary/#pii) [Synthetic Data](https://www.jkconcepts.at/en/ai-glossary/#synthetic-data) ## Terms starting with Q Development · Technical [#quantization](https://www.jkconcepts.at/en/ai-glossary/#quantization) ### What does quantisation mean in AI? Quantisation · Reducing a model’s numerical precision Reducing the numerical precision of a model to lower memory requirements and computing effort. ### Technology, example and context #### Example A quantised language model runs on a smaller graphics card or a powerful laptop. #### Why does it matter? Quantisation can enable local and cheaper applications, but depending on the method it can cost some quality. #### Related terms [Local AI / Local LLM](https://www.jkconcepts.at/en/ai-glossary/#local-ai) [GPU & VRAM](https://www.jkconcepts.at/en/ai-glossary/#gpu-vram) [Parameters](https://www.jkconcepts.at/en/ai-glossary/#parameters) ## Terms starting with R Language models · Advanced [#rag](https://www.jkconcepts.at/en/ai-glossary/#rag) ### What is RAG (retrieval-augmented generation)? RAG · Retrieval-Augmented Generation A method in which an AI model retrieves suitable information from defined sources before answering. ### Technology, example and context #### Example An internal search finds relevant product documentation; the model formulates an answer from it with source references. #### Why does it matter? RAG can make answers more current and traceable if retrieval, permissions and source quality are implemented properly. #### Related terms [Retrieval](https://www.jkconcepts.at/en/ai-glossary/#retrieval) [Embedding](https://www.jkconcepts.at/en/ai-glossary/#embedding) [Vector Database](https://www.jkconcepts.at/en/ai-glossary/#vector-database) [Grounding](https://www.jkconcepts.at/en/ai-glossary/#grounding) Data · Advanced [#retrieval](https://www.jkconcepts.at/en/ai-glossary/#retrieval) ### What does Retrieval mean? Retrieval · Information retrieval Targeted finding of relevant information before a system produces an answer or decision. ### Technology, example and context #### Example From thousands of document sections, the five most likely to match a query are selected. #### Why does it matter? Poor retrieval gives even a strong model the wrong basis; search quality and permissions are therefore central. #### Related terms [RAG](https://www.jkconcepts.at/en/ai-glossary/#rag) [Semantic Search](https://www.jkconcepts.at/en/ai-glossary/#semantic-search) [Knowledge Base](https://www.jkconcepts.at/en/ai-glossary/#knowledge-base) ## Terms starting with S AI search · Advanced [#schema-org](https://www.jkconcepts.at/en/ai-glossary/#schema-org) ### What does Schema.org mean? Schema.org · Vocabulary for structured web data A shared vocabulary that websites use to describe people, organisations, offers, terms and other content. ### Technology, example and context #### Example A glossary uses DefinedTermSet and DefinedTerm to mark up terms and definitions in a machine-readable way. #### Why does it matter? Schema.org creates a common language for systems; markup alone does not replace helpful visible content, however. #### Related terms [Structured Data](https://www.jkconcepts.at/en/ai-glossary/#structured-data) [Entity](https://www.jkconcepts.at/en/ai-glossary/#entity) [GEO](https://www.jkconcepts.at/en/ai-glossary/#geo) Data · Advanced [#semantic-search](https://www.jkconcepts.at/en/ai-glossary/#semantic-search) ### What is semantic search? Semantic Search · Search by meaning Search by meaning and intent rather than only by exactly matching words. ### Technology, example and context #### Example A search for “correct an invoice” also finds a guide titled “Create a credit note”. #### Why does it matter? Semantic search improves findability with different wording, but should be combinable with filters and classic search. #### Related terms [Embedding](https://www.jkconcepts.at/en/ai-glossary/#embedding) [Retrieval](https://www.jkconcepts.at/en/ai-glossary/#retrieval) [Vector Database](https://www.jkconcepts.at/en/ai-glossary/#vector-database) Basics · Basic [#speech-to-text](https://www.jkconcepts.at/en/ai-glossary/#speech-to-text) ### What does Speech-to-Text mean? Speech-to-Text · Automatic speech recognition Converting spoken language into written text. ### Technology, example and context #### Example A meeting recording is transcribed and then searched for decisions. #### Why does it matter? The technology makes documentation and search easier, but confidential conversations need suitable data protection and retention rules. #### Related terms [Text-to-Speech](https://www.jkconcepts.at/en/ai-glossary/#text-to-speech) [Multimodal AI](https://www.jkconcepts.at/en/ai-glossary/#multimodal-ai) [Data Protection](https://www.jkconcepts.at/en/ai-glossary/#datenschutz) AI search · Advanced [#structured-data](https://www.jkconcepts.at/en/ai-glossary/#structured-data) ### What is structured data? Structured Data · Machine-readable page data Machine-readable information that describes the content and meaning of a page in defined fields. ### Technology, example and context #### Example A website additionally marks up its organisation, services or defined terms as JSON-LD. #### Why does it matter? Structured data makes unambiguous interpretation easier, but must match the visible content and be technically correct. #### Related terms [Schema.org](https://www.jkconcepts.at/en/ai-glossary/#schema-org) [Entity](https://www.jkconcepts.at/en/ai-glossary/#entity) [Knowledge Graph](https://www.jkconcepts.at/en/ai-glossary/#knowledge-graph) Language models · Advanced [#structured-output](https://www.jkconcepts.at/en/ai-glossary/#structured-output) ### What does Structured Output mean? Structured Output · Output in a defined format Model output that follows a predefined format or data schema. ### Technology, example and context #### Example Instead of free text, the model returns fields such as category, priority and reasoning as validatable JSON. #### Why does it matter? Structured outputs can be processed further more safely, but still have to be validated technically and checked for content. #### Related terms [Tool & Function Calling](https://www.jkconcepts.at/en/ai-glossary/#tool-calling) [API](https://www.jkconcepts.at/en/ai-glossary/#api) [AI Automation](https://www.jkconcepts.at/en/ai-glossary/#ai-automation) Data · Advanced [#synthetic-data](https://www.jkconcepts.at/en/ai-glossary/#synthetic-data) ### What does Synthetic Data mean? Synthetic Data · Artificially generated data Artificially generated data intended to mimic real structures or cases. ### Technology, example and context #### Example For a prototype, invented customer cases are created with typical but not real personal data. #### Why does it matter? Synthetic data can make testing easier and reduce data protection risks, but it does not automatically reproduce rare or problematic real cases correctly. #### Related terms [Training Data](https://www.jkconcepts.at/en/ai-glossary/#training-data) [Data Protection](https://www.jkconcepts.at/en/ai-glossary/#datenschutz) [Evaluation, Evals & Benchmarks](https://www.jkconcepts.at/en/ai-glossary/#evaluation-evals) Language models · Advanced [#system-prompt](https://www.jkconcepts.at/en/ai-glossary/#system-prompt) ### What does System Prompt mean? System Prompt · Top-level model instruction A top-level instruction that defines a model’s role, rules and desired behaviour. ### Technology, example and context #### Example The system prompt requires the model to answer only from approved documents and to state uncertainty explicitly. #### Why does it matter? System prompts create guardrails, but are no reliable protection against attacks or incorrect outputs. #### Related terms [Prompt](https://www.jkconcepts.at/en/ai-glossary/#prompt) [AI Guardrails](https://www.jkconcepts.at/en/ai-glossary/#ai-guardrails) [Prompt Injection](https://www.jkconcepts.at/en/ai-glossary/#prompt-injection) ## Terms starting with T Language models · Advanced [#temperature](https://www.jkconcepts.at/en/ai-glossary/#temperature) ### What does Temperature mean? Temperature · Randomness setting of the output A setting that, in many models, influences how predictable or varied an output is. ### Technology, example and context #### Example A lower temperature is usually chosen for data extraction, a higher one possibly for brainstorming. #### Why does it matter? The setting changes how choices are made, but makes an answer neither automatically correct nor fully reproducible. #### Related terms [Inference](https://www.jkconcepts.at/en/ai-glossary/#inference) [Structured Output](https://www.jkconcepts.at/en/ai-glossary/#structured-output) [Evaluation, Evals & Benchmarks](https://www.jkconcepts.at/en/ai-glossary/#evaluation-evals) Basics · Basic [#text-to-speech](https://www.jkconcepts.at/en/ai-glossary/#text-to-speech) ### What does Text-to-Speech mean? Text-to-Speech · Speech synthesis Generating spoken language from written text. ### Technology, example and context #### Example An application reads out instructions in several languages, making them more accessible. #### Why does it matter? Text-to-speech improves accessibility and usability, but cloned voices raise additional questions of consent and misuse. #### Related terms [Speech-to-Text](https://www.jkconcepts.at/en/ai-glossary/#speech-to-text) [Generative AI](https://www.jkconcepts.at/en/ai-glossary/#generative-ai) [AI Guardrails](https://www.jkconcepts.at/en/ai-glossary/#ai-guardrails) Language models · Advanced [#token](https://www.jkconcepts.at/en/ai-glossary/#token) ### What are tokens in AI? Token · A model’s unit of text A small unit into which a model splits text; a token can be a word, part of a word or a punctuation mark. ### Technology, example and context #### Example A long German compound word can consist of several tokens and therefore use up more context space. #### Why does it matter? Token counts affect context limits, processing costs and response times of many AI services. #### Related terms [Tokenisation](https://www.jkconcepts.at/en/ai-glossary/#tokenization) [Context Window](https://www.jkconcepts.at/en/ai-glossary/#context-window) [Latency](https://www.jkconcepts.at/en/ai-glossary/#latency) Language models · Advanced [#tokenization](https://www.jkconcepts.at/en/ai-glossary/#tokenization) ### What does Tokenisation mean? Tokenisation · Splitting text into tokens Splitting an input into tokens that the model can process. ### Technology, example and context #### Example Before the model processes a text, it is broken down into words, parts of words and punctuation marks. #### Why does it matter? Different models count the same text differently; this matters for limits, costs and multilingual content. #### Related terms [Token](https://www.jkconcepts.at/en/ai-glossary/#token) [Context Window](https://www.jkconcepts.at/en/ai-glossary/#context-window) [LLM](https://www.jkconcepts.at/en/ai-glossary/#llm) Agents & automation · Advanced [#tool-calling](https://www.jkconcepts.at/en/ai-glossary/#tool-calling) ### What does Tool & Function Calling mean? Tool & Function Calling · Structured tool invocation A model’s ability to propose a structured call to an external function or piece of software. ### Technology, example and context #### Example The model generates a validated call to retrieve a delivery status from a business system. #### Why does it matter? Tool calling connects language with real actions; permissions, input validation and confirmation of critical steps remain the job of the application. #### Related terms [AI Agent](https://www.jkconcepts.at/en/ai-glossary/#ai-agent) [API](https://www.jkconcepts.at/en/ai-glossary/#api) [Structured Output](https://www.jkconcepts.at/en/ai-glossary/#structured-output) Language models · Advanced [#training](https://www.jkconcepts.at/en/ai-glossary/#training) ### What does Training mean? Training · Model training The process in which a model adjusts its internal parameters using many examples. ### Technology, example and context #### Example An image model sees numerous labelled examples and learns to distinguish certain product defects. #### Why does it matter? Training a model from scratch is too costly for most businesses; existing models, RAG or limited adaptation are often enough. #### Related terms [Training Data](https://www.jkconcepts.at/en/ai-glossary/#training-data) [Parameters](https://www.jkconcepts.at/en/ai-glossary/#parameters) [Fine-Tuning](https://www.jkconcepts.at/en/ai-glossary/#fine-tuning) Data · Advanced [#training-data](https://www.jkconcepts.at/en/ai-glossary/#training-data) ### What does Training Data mean? Training Data · Data a model learns from Data from which a model learns patterns and relationships during training. ### Technology, example and context #### Example Historical, correctly categorised enquiries serve as examples for later automatic assignment. #### Why does it matter? Gaps, errors or biases in training data can show up directly in the later results. #### Related terms [Training](https://www.jkconcepts.at/en/ai-glossary/#training) [Bias](https://www.jkconcepts.at/en/ai-glossary/#bias) [Model Drift](https://www.jkconcepts.at/en/ai-glossary/#model-drift) Language models · Advanced [#transformer](https://www.jkconcepts.at/en/ai-glossary/#transformer) ### What is a transformer model? Transformer · Transformer architecture A model architecture that weights the relationships between elements of an input and enables many modern language models. ### Technology, example and context #### Example When processing a paragraph, the model takes into account which words matter for the meaning of other words. #### Why does it matter? Transformers explain part of the power of modern models, but on their own say nothing about data quality, cost or reliability. #### Related terms [LLM](https://www.jkconcepts.at/en/ai-glossary/#llm) [Neural Network](https://www.jkconcepts.at/en/ai-glossary/#neural-network) [Context Window](https://www.jkconcepts.at/en/ai-glossary/#context-window) ## Terms starting with V Data · Advanced [#vector](https://www.jkconcepts.at/en/ai-glossary/#vector) ### What does Vector mean? Vector · Numerical feature vector An ordered list of numbers with which an ML system represents properties of content mathematically. ### Technology, example and context #### Example The vector of a text lies closer to texts with similar content than to unrelated ones. #### Why does it matter? Vectors are the technical form behind embeddings and similarity search, but are not directly interpretable by humans. #### Related terms [Embedding](https://www.jkconcepts.at/en/ai-glossary/#embedding) [Vector Database](https://www.jkconcepts.at/en/ai-glossary/#vector-database) [Semantic Search](https://www.jkconcepts.at/en/ai-glossary/#semantic-search) Data · Advanced [#vector-database](https://www.jkconcepts.at/en/ai-glossary/#vector-database) ### What is a vector database? Vector Database · Database for similarity search A database or search system that stores vectors and efficiently finds similar entries. ### Technology, example and context #### Example Product texts are stored as embeddings so that a question finds matching sections by meaning. #### Why does it matter? A vector database is often part of RAG, but on its own solves neither content quality, updating nor access rights. #### Related terms [Embedding](https://www.jkconcepts.at/en/ai-glossary/#embedding) [RAG](https://www.jkconcepts.at/en/ai-glossary/#rag) [Semantic Search](https://www.jkconcepts.at/en/ai-glossary/#semantic-search) ## Terms starting with W Agents & automation · Advanced [#workflow-automation](https://www.jkconcepts.at/en/ai-glossary/#workflow-automation) ### What does Workflow Automation mean? Workflow Automation · Process automation Automating a predefined process with defined steps and rules. ### Technology, example and context #### Example A validated form triggers filing, a notification and the creation of a task, one after the other. #### Why does it matter? Classic workflows are often more traceable than autonomous agents and the better choice for stable, rule-based processes. #### Related terms [AI Automation](https://www.jkconcepts.at/en/ai-glossary/#ai-automation) [Orchestration](https://www.jkconcepts.at/en/ai-glossary/#orchestration) [API](https://www.jkconcepts.at/en/ai-glossary/#api) ## Terms starting with Z Language models · Advanced [#zero-shot](https://www.jkconcepts.at/en/ai-glossary/#zero-shot) ### What does Zero-Shot mean? Zero-Shot · Task without examples A model is expected to solve a task from an instruction alone, without first receiving suitable examples. ### Technology, example and context #### Example The model classifies an enquiry as sales or support even though the prompt shows no sample cases. #### Why does it matter? Zero-shot is quick to set up, but can be less consistent with company-specific categories. #### Related terms [Few-Shot](https://www.jkconcepts.at/en/ai-glossary/#few-shot) [Prompt Engineering](https://www.jkconcepts.at/en/ai-glossary/#prompt-engineering) [Evaluation, Evals & Benchmarks](https://www.jkconcepts.at/en/ai-glossary/#evaluation-evals) From definition to application ## Integrating AI sensibly into your website or processes? Not every business needs a chatbot. What matters is the task, the data, the benefit, the risks and an operation that stays under control.