ChatGPT Enterprise, on Casablanca time

A rollout rarely begins with the technology. Our method starts with a short scoping exercise: which business functions, which use cases, which data are allowed to pass through it. We map the data flows — indispensable for the CNDP file — then configure the workspace: SSO against your directory, groups and permissions, internal GPTs connected to your document repositories, usage rules published in French and in Arabic where needed.

Hunter BI is a member of the OpenAI partner network and supports you through the licensing contract — billed in foreign currency, a point to anticipate with your finance department in light of exchange-control regulations — then remains your local point of contact after go-live. Adoption is measured in the administration console: activation rate per team, emerging use cases and further training needs.

This is the question that decides everything for Moroccan service providers: a contact centre or a BPO that processes the data of European citizens acts as a processor within the meaning of the GDPR, while remaining subject to Law 09-08 on the Moroccan side. Introducing ChatGPT Enterprise into the processing chain therefore calls for a dual reading: a data processing agreement with OpenAI, notification of the European data controller, a CNDP declaration or authorisation depending on the nature of the data, and minimisation upstream — customer identifiers most often have no reason to enter a prompt.

ChatGPT Enterprise provides contractual and technical building blocks — a commitment not to train on your data, encryption, SOC 2 certification, and data-residency options announced by OpenAI for certain regions. Our role is to document their configuration and the data flows in a file your clients can audit. Your legal and compliance owners then validate whether the arrangement fits your situation.

Hunter BI is a member of the OpenAI partner network. That status does not replace the work specific to your project: inventorying the offers actually available, sizing licences, validating the contract and checking features in your workspace.

Our support covers three documented decisions. Sizing: choosing the tier from the number of users and intended uses. Architecture: arbitrating between native functions, connectors and the API on total cost. Evolution: revisiting the roadmap when capabilities or commercial terms change, without promising a feature before its availability is verified.

A successful ChatGPT Enterprise rollout in Morocco follows a proven cadence. Weeks 1 to 2: scoping, data mapping, usage rules. Weeks 3 to 6: a pilot with two or three volunteer teams — often customer relations, support functions and one business team — with measurable use cases: replying to multilingual emails in FR/AR/EN, summarising case files, preparing deliverables. Weeks 7 to 10: rollout in waves, training by function, internal GPTs built on your procedures.

Two pitfalls recur systematically in the deployments we take over: licences handed out without use cases — the tool dies within six weeks — and a blanket ban that pushes teams back towards their personal accounts, that is, towards shadow AI. The answer to both is the same: defined, measured and equipped uses, championed by trained referents in every department.