In most companies, AI training starts the same way: a meeting room, a slide deck and two hours titled "What you can do with ChatGPT." The following week, everyone goes back to their old habits. The problem isn't the content; it's teaching everyone the same thing. Someone in accounting needs something different from the marketing team, and the decisions a manager has to make are another matter entirely.
This post splits training by role, helps you plan the content and length of each level, and gives separate space to the two topics that get skipped most often: data rules and measurement.
A significant share of employees already use AI tools, often with personal accounts and without asking anyone. That isn't bad in itself, but it carries two risks: customer or company data ending up in an uncontrolled tool, and wrong information from the tool being used without being checked.
Two official texts bear directly on this:
Three role-based levels
Safe use and core skills
Which data can go in and which can't; how to ask a good question; how to check the answer.
Decisions and rules
Which work it will be used for, which tools are approved, who is responsible.
In-depth use in their own work
Workflows, templates and measurable goals tailored to the team's daily work.
The goal of this level is for everyone to know the same basic rules:
Managers need to make decisions more than they need to use the tool:
The answers to these questions become a short, written usage policy. The draft should be ready before the training; Level 1 training explains that policy.
Here, the training changes with each team's work. Content drafts and image generation for marketing, proposals and emails for sales, spreadsheets and reports for accounting, reply templates for customer service. The output of this level shouldn't be a presentation; it should be a few templates and workflows the team will use every day.
| Module | Level 1 | Level 2 | Level 3 |
|---|---|---|---|
| What AI can and can't do | Yes | Yes | Short recap |
| Data rules and KVKK | Yes | Yes, in detail | Yes |
| Asking good questions and checking | Yes | Short | Yes, in depth |
| Usage policy | Explained | Written | Applied |
| Tool selection and accounts | No | Yes | Short |
| Practice with their own work | Short | No | Yes, main part |
| Measurement and goals | No | Yes | Yes |
The length depends on the size of the company, the number of teams and the starting level. The plan below is a planning assumption, not a measured average; use it as a starting point when you build your own schedule.
A satisfaction survey at the end of the training doesn't show that the training worked. Instead, look at these one month later:
Common practice
What works
Is my training plan ready?
0/6Compare the six items with your own plan; every item you don't check is a job to do before the training.
In our AI consulting service, before any training we look at how teams use AI today, write the draft usage policy together with managers, then split the training by role and have each team apply it to its own work. One month later, we review the measurements together. I also covered where to start in a company in the AI transformation guide.
In Turkey, as of November 2026, no separate law regulating AI is in force; the Artificial Intelligence Action Plan (2026-2030) circular published in August 2026 is not a law but an administrative regulation. Because legislation can change, check the current situation with your lawyer. Article 4 of the EU AI Act, on the other hand, requires organizations that use AI systems in the EU to take measures for a sufficient level of AI literacy among their staff. If you serve the EU, this article may concern you; check with your lawyer.
A ban usually doesn't eliminate usage; it just makes it invisible. An approved tool, a company account and a clear data rule are generally safer. I covered this question in detail in the AI transformation guide.
For Level 1 and Level 3, yes, especially if there's an employee who already uses the tools well. At Level 2, meaning the policy and data rules part, the legal and IT side also need to be at the table.
For practice, yes, but how data is used on free and personal accounts can differ from business accounts. If employees will practice with their own real work during training, do it on the company's approved account.
Look at the measurements one month later, put the templates that work in the team's shared folder, and update the policy based on real usage. Since AI tools change quickly, plan a short refresher session every few months.
AI projects often either never get started or get stuck in the pilot. Here is a 90-day plan split into three phases, discovery, pilot and rollout, with the deliverables you should have in hand at the end of each one.
According to Comscore's September 22, 2026 report, ChatGPT's share of AI assistant conversations fell from 70% to 50% in six months; the same day, Claude Opus 5.5 launched at a lower price. The risk of an AI setup tied to a single vendor, and what to do today.
October 4, 2026: starting October 9, Google is limiting the Gemini app on free personal accounts to the Flash-Lite model only, and Pro is also going away on AI Plus.
We use cookies to improve the site and measure advertising performance. Analytics and advertising cookies run only if you accept them. Details are in our Privacy Policy and Data Protection Notice.
