AI training for your team

When training helps

Many teams, from small businesses to departments in larger companies, have access to AI tools that haven't changed how work gets done. A few people use them every day, others tried once, and nobody is sure what's allowed. Some common signs, as examples:

• Each person writes their own prompts each time, and results vary from person to person.
• Nobody has agreed which tasks are fine to hand to AI and which need a person.
• Good results depend on one colleague who knows what to type.
• People worry about putting client information into the wrong tool.

How a session works

Each session is scoped to the people and tools involved. We work through four steps:

1. Bring a real recurring task. Something your team does every week, such as drafting a client update or summarizing notes from a call. Those are examples; your team picks the task.
2. Practice with your existing tools. Everyone works on the task with the AI tools your business already uses or has approved.
3. Document the working process. We write down the steps, the inputs the task needs, and what a good result looks like.
4. Agree when a person steps in. The team decides which outputs need a check, who does it, and what happens when something looks wrong.

Working on your own task shows quickly where AI helps with it and where it falls short.

What your team keeps

• A practiced workflow for the task you brought.
• A written reference your team can return to.
• Where it helps, reusable prompts or a playbook with examples and review criteria.
• Agreed rules for when a person reviews the output.

Human review and your data

Your team checks accuracy and fit before using AI output. The review rules are written into the documented workflow so they don't depend on memory. Which tools are approved, and what information can go into them, is agreed before the session.

Limits to know about

Some tasks are a poor fit for AI because they are too varied, too sensitive, or depend too heavily on judgment, and we'll tell you when that's the case. New habits also take practice after the session, which is what the written reference is for. If a task would be better handled by a built automation than by a prompt, that becomes a separate project, and the AI Leverage Audit is the way to scope it.

Who teaches it

Common Ground's training is led by a developer who has taught programming to students ranging from kids to working professionals, and who has spent three years shipping production software. That mix helps judge whether a task is better handled with a prompt or with a built system.

Where to start

The free AI readiness checklist takes about 10 minutes. One of its questions asks whether someone on your team has used an AI tool for real work in the past month, which is a useful starting point for a training conversation. Questions about data and how the work happens are answered on the FAQ page.