Daniel & Friends / Glossary / AI safety and alignment
AI and automationAI safety and alignment
AI safety is the field concerned with making artificial intelligence systems behave safely and predictably. Alignment is part of it: the effort to match a model’s behavior with human intent and values so that it does what users actually want.
Daniel Votruba ·
In practice
For a company this means choosing vendors that document their safety testing and deploying AI first where the impact of a mistake is small and easy to fix.
Related terms
- Guardrails: Guardrails are technical and procedural rules that monitor the inputs and outputs of an AI system.
- AI governance: AI governance is the set of rules, roles and processes a company uses to manage its use of artificial intelligence.
- 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.
- Hallucination: A hallucination is a case where a model states plausible-sounding but false information.
- EU AI Act: The EU AI Act is the European regulation on artificial intelligence.
More terms in the group AI and automation
AI scoring · AI personalization · AI SDR · AI search (answer engines) · AI crawlers (GPTBot and others) · AI slop · AI costs and limits · AI workflow and orchestration · AI automation · Automation platforms (Zapier, Make, n8n) · AI agent · C2PA and content provenance (AI watermarks) · Agentic AI · Chain of thought · Voice cloning · ChatGPT Ads · Vibe coding · Computer vision · Vector database · Context window · Token · Copilot (AI assistant) · Text-to-video · Custom GPTs and projects
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