GLOSSARY

The language of AI capability.

Clear definitions for the words that get used loosely: capability, upskilling, adoption, literacy, and what the rules actually mean. Each links to a fuller answer.

AI capability

AI capability system: An AI capability system is software that builds and keeps a record of how well a person and a team can actually use AI in their real work.
System of record for AI capability: A system of record for AI capability is the single, trusted place where an organisation's real AI ability is measured and kept.
AI training versus AI capability: AI training is an event.
Building AI capability in a team: You build AI capability in a team by turning AI use into a daily habit tied to real work, measuring the before and after on actual tasks, and keeping a shared record that managers can see.
Keeping AI skills current: You keep AI skills current by tying them to practice, not to a fixed curriculum.
AI fluency: AI fluency is the point where using AI in your work stops being an effort and becomes second nature.
AI certifications and capability: Not on their own.

Upskilling and reskilling

Upskilling a team for AI: You upskill a team for AI by building the skill into daily work rather than into a one-off course.
Upskilling versus reskilling: Upskilling means getting better at the job you already have.
AI upskilling program: An AI upskilling program is a structured effort to raise a workforce's ability to use AI.
AI skills the workforce needs: The workforce needs practical, role-specific AI skills more than technical ones: framing a task for AI, steering and refining its output, judging when to trust it, and knowing where a human must stay in charge.
How long AI upskilling takes: A first useful win can happen in the first session, often within minutes on a real task.
Reskilling for AI-driven change: You reskill workers whose jobs are changing by moving them onto the parts of the work AI cannot do well, and building that new capability through real practice with support.
AI skills gap: The AI skills gap is the distance between the AI ability an organisation has and the ability its work now demands.

AI adoption

Software tourism: Software tourism is when people visit a new tool, look around, and go back to their old way of working.
Why AI pilots fail: Most corporate AI pilots fail because they are run as events, not habits.
AI adoption strategy: An AI adoption strategy is a plan for turning AI from a few experiments into everyday capability across an organisation.
From AI pilot to scale: You move from a pilot to scale by proving capability in the pilot, then repeating the same daily loop team by team.
Measuring AI adoption: Measure AI adoption by depth of use, not licences issued.
Getting a team to use AI tools: You get a team to actually use AI tools by tying them to work people already do and dislike, giving one clear action a day, and showing the result.

Learning and L&D

LMS versus AI capability system: A learning management system stores courses and tracks who completed them.
Why AI courses do not stick: AI courses don't stick because they deliver information once and then send people back to unchanged work.
AI training versus AI enablement: AI training teaches people about AI.
AI in corporate learning: Learning and development teams should treat AI as a capability to build in the flow of work, not a course to deliver.
AI literacy: AI literacy is a person's ability to understand what AI can and cannot do, use it responsibly, and judge its output in their own context.

HR and measurement

Measuring AI capability: You measure AI capability by looking at real work, not quizzes or course completion.
Proving AI capability to leadership: You prove AI capability to leadership with evidence from real work, not activity reports.
AI skills matrix: An AI skills matrix is a simple map of who on a team can do what with AI, and how well.
AI capability score: An AI capability score is a summary of how well a person or team can use AI in their real work, built from evidence rather than opinion.
Benchmarking AI ability: You benchmark a team's AI ability by measuring the same real-work signals consistently across people and over time: which tasks each person can improve with AI, whether the gains are verified, and how capability trends month to month.

EU AI Act and compliance

AI literacy under the EU AI Act: Since 2 February 2025, the EU AI Act requires providers and deployers of AI systems to make sure the people who use AI on their behalf have a sufficient level of AI literacy.
EU AI Act enforcement timeline: The EU AI Act applies in stages.
Scope of EU AI Act Article 4: Article 4 applies to providers and deployers of AI systems.
Sufficient AI literacy: The EU AI Act does not set a fixed definition of sufficient.
Complying with the EU AI Act AI literacy duty: You comply with the EU AI Act AI literacy requirement by making sure the people who use AI on your behalf understand it well enough for their role, and by keeping evidence that you did so.

Read the full answers →See ARK by country →