Use AI like you mean it.

The one shift that changes everything

Most people get stuck because they treat AI like a search engine: type a few words, expect a perfect answer, get something generic, give up. That's the wrong mental model.

Treat AI like a brilliant new hire who has read almost everything — but knows nothing about your company, your customers, or what you actually need today.

That one reframe fixes nearly everything. A sharp new hire is enormously capable, but on day one you wouldn't hand them a one-line instruction and walk away. You'd give them context, show them an example, tell them what "good" looks like, and check their work before it goes out the door. AI is exactly the same.

Three things follow from this, and they run through the rest of this guide: give it context (it can't read your mind), review its work (capable doesn't mean correct), and stay in charge (it assists; you decide).

The anatomy of a great prompt

You don't need to memorize tricks. Almost every strong prompt has the same five ingredients. Include even three of them and your results jump.

Watch the difference one set of ingredients makes:

Same model, same thirty seconds of effort — wildly different output. The second one you can send; the first one you have to rewrite.

The five moves

Strip away the buzzwords and almost everything useful AI does at work is one of five moves. Learn to spot which one you need and you'll always know how to start.

When you're stuck on how to use AI for a task, ask: which of the five moves is this? Then add your ingredients from section two.

Work the way you actually work

The fastest wins come from pointing AI at the boring, repetitive parts of your existing job — not from inventing new work. A few patterns by role to spark ideas:

If you're in sales

Prep for calls from a prospect's website, turn call transcripts into summaries and next steps, and draft follow-ups that reference the actual conversation.

If you're in marketing

Turn one idea into a week of content in your voice, generate headline and angle options to react to, and draft briefs from rough notes.

If you're in ops or admin

Convert messy notes into clean action plans, draft SOPs from a quick description, and triage a full inbox into what matters now.

If you lead a team

Draft decision memos, prep for 1:1s, and find a clear, kind way to say the hard thing.

Every one of these — ready to copy — lives in your Prompt Library. Start there when you want a running start.

Trust, but verify

AI is fluent, which makes it easy to believe — and it can be fluent and wrong at the same time. Confidence is not accuracy. Before you rely on any output, run the four-question check:

• Did I check the facts, names, and numbers?
• Did I keep confidential or personal data out of it?
• Would I stand behind this if it went out under our name?
• Is a person — not the tool — owning the final decision?

If any answer gives you pause, fix that before you send. The full policy and an interactive version of this check live on your Governance & Safety page.

Make it a habit

Skills you don't use evaporate. The teams that get fluent aren't the ones that took a course — they're the ones who built a small daily habit. Here's how to make it stick:

Keep a personal prompt file. Every time a prompt works, paste it somewhere you'll find it again. In a month you'll have your own library worth more than any course.

Add one reflex. At the start of any task, ask: "Could AI give me a usable first draft of this?" Often the answer is yes, and you've just saved twenty minutes.

Practice for two minutes a day. The games in this portal are built for exactly this — quick reps that train your eye for what makes a prompt work.

Fluency isn't about knowing more tricks. It's about reaching for AI without thinking — and knowing when not to.

You don't need to be an expert. You need to be the kind of person who reaches for the right tool, gives it what it needs, checks the result, and owns the call. That's the whole game. Now go use it.