In recent years, AI vocabulary has filled up with words that impress more than they explain. Here are the most useful terms to understand, explained simply.
Custom-built tool
A tool designed specifically for a company's business, rather than off-the-shelf software sold the same way to everyone. This is the opposite of a "licensed" or "standard" tool, already built and identical for every customer.
Audit
An analysis phase, usually billed, that identifies tasks that can be automated within a company and estimates the expected gain before any decision to build a tool. A serious audit can conclude that nothing should be built.
GDPR
The General Data Protection Regulation, in force since 2018 across the European Union. It governs the collection, use, and storage of personal data: a topic directly relevant to any AI use that processes customer or employee data.
AI Act
The European regulation on artificial intelligence, rolled out in successive stages since 2025. It classifies AI uses by risk level and imposes obligations (transparency, training, documentation) proportionate to that risk level.
AI-powered social media management
The automated creation and publishing of branded content (visuals, videos, text) on social media, based on a company's visual identity and tone of voice. This is what Isalis does.
Compliance register
A document that lists every AI system used by a company, along with its risk level and compliance status. It serves as a reference in the event of an audit and as a basis for any future decision.
Provider vs. software vendor
A software vendor sells the same software to all its customers. A provider that builds custom solutions designs a different tool for each company, tailored to its specific business.
White label
An offering that an agency or firm can resell to its own clients under its own name, without the end client knowing who actually built the tool behind the scenes.
Write to us, we are happy to expand this glossary.
