Daniel & Friends / Glossary / Explainable AI (XAI)
AI and automationExplainable AI (XAI)
Explainable AI (XAI) is the ability to show in understandable terms why a model reached a given result: which inputs mattered, how confident it is and what the answer rests on. For large language models this is still only partly achievable.
Daniel Votruba ·
In practice
Demand explainability where decisions affect people or money. For text outputs it helps when the model cites sources and reasoning that you can verify.
Related terms
- Model bias (AI bias): Model bias is a systematic distortion in AI outputs that comes from unbalanced training data, model design or the way a task is framed.
- AI governance: AI governance is the set of rules, roles and processes a company uses to manage its use of artificial intelligence.
- EU AI Act: The EU AI Act is the European regulation on artificial intelligence.
- Hallucination: A hallucination is a case where a model states plausible-sounding but false information.
- RAG (retrieval-augmented generation): RAG is a method in which a model first retrieves relevant documents, for example company knowledge, and answers from them.
More terms in the group AI and automation
Few-shot and zero-shot prompting · Fine-tuning · Embedding · Foundation model · Diffusion models · Function calling (tool use) · Deep learning · Generative AI · Deepfake and synthetic media · Grounding · Custom GPTs and projects · Guardrails · Copilot (AI assistant) · Context window · Human in the loop · Computer vision · Inference and latency · ChatGPT Ads · Jailbreak (AI) · Chain of thought · Large language model (LLM) · C2PA and content provenance (AI watermarks) · Machine learning (ML) · Automation platforms (Zapier, Make, n8n)
Terms in practice
Need help putting them to work?
Terms are one thing, putting them to work is another. We go through which of them pay off for you and say so plainly. You get a first opinion for free and we reply within 24 hours.
Book a call