The right information, at the right time. AI Studio queries your documents and the web securely.

Studio IA works in 4 automatic steps as soon as you upload a file. Think of it as an ultra-fast librarian who: (1) reads your document, (2) splits it into logical paragraphs ("chunks"), (3) creates a vector fingerprint of each paragraph to understand its meaning — not just its keywords —, then (4) indexes everything into a searchable index accessible in milliseconds. Supported formats include PDF, Word (.docx), Excel (.xlsx), PowerPoint, emails (.eml, .msg), Markdown, HTML, CSV and 20+ others. No preparation or conversion needed. Indexing takes 2 to 5 minutes for a standard 50-page document.

Unlike ChatGPT which generates answers from memory, Studio IA systematically cites its source for every element of a response: the document name, page number, and exact section. One click opens the source passage for verification. It's like the difference between a colleague who says "I think the clause is in the contract" and one who says "it's on page 7, clause 3.2, here's the exact text." For accounting or legal firms, this traceability is indispensable.

Yes. With the sovereign GPU option (Pro and Enterprise plans), all processing — ingestion, vectorization, LLM inference — happens on servers physically located in your country. Neither your documents nor your queries transit to OpenAI, Google, or any foreign provider. For Starter plans using an external LLM, only the relevant passages ("chunks") are sent — never the full document — and personal data is masked before transmission.

The Starter plan supports 3 to 5 users, Pro 5 to 25, and Enterprise is unlimited. Simultaneous collaboration is supported without performance degradation thanks to the multi-tenant architecture. In practice: a firm of 12 accountants can all query the same document base simultaneously during tax season, with a target latency of under 3 seconds.

Three options: (1) manual update — re-upload the file, the old version is automatically replaced, (2) automatic sync via connector (Google Drive, SharePoint, OneDrive) — any new file in a designated folder is indexed within 15 minutes, (3) API for systems that generate documents programmatically. Example: a clinic that updates its protocols quarterly sets up a SharePoint sync — as soon as the medical director saves the updated document, the team queries the latest version with no technical intervention.

Yes. Studio IA supports separate workspaces by department and fine-grained document permissions. An HR assistant can access internal policies without seeing financial records. Access management is configured from the admin panel without technical skills, using a logic identical to Google Drive or SharePoint.

A classic chatbot uses keyword search: it finds documents that "contain the words" in your question. Studio IA uses vector semantic search: it understands the meaning of your question, even if the exact words aren't in the document. Example: you ask "what are our usual payment terms?" — a chatbot finds nothing if your contracts say "NET 30" without writing "payment terms." Studio IA understands that NET 30 is a payment condition. Furthermore, it synthesizes information from multiple documents into a coherent answer.