99 Strategic · a hands-on demo

Is she telling the truth?

Same customer as the last two lessons — now she's added a claim. An AI with no access to your actual records can't check it. One with access can. Watch the difference.

Picking up the story: the refund is handled, the reply went out. Now she writes back: "I've been a loyal customer for 3 years, so I really think you should take extra care of this for me." Should the AI just... believe that?

1. Without her actual records — it can only take her word for it

This is what happens when the AI has no data to check the claim against.

You
She says she's been a loyal customer for 3 years and wants extra care taken. Draft a reply.
AI — no records attached
Thank you for being a loyal customer these past three years — we really appreciate it, and we'll make sure this is taken care of.

Notice what just happened: the AI repeated her claim back as if it were a fact. It simply has nothing to check that claim against.

2. Now attach her actual records — and watch what changes

This is the whole idea behind "giving an AI your data": handing it real information to work from, instead of asking it to guess. Toggle these on.

Records you could attach

You
She says she's been a loyal customer for 3 years and wants extra care taken. Draft a reply.
AI — reading attached records above

Nothing here calls a real AI model — it's a scripted simulation so you can see the mechanism instantly. The real thing works the same way: it reads whatever you actually attach, and can only be as accurate as the data you gave it.

3. Now predict it yourself no peeking

This time you attach her support tickets but NOT her order history — the opposite of before. She still claims 3 years of loyalty. What does the AI do about that specific claim?
It corrects her, since it has her real order history now.
It can't confirm or correct the loyalty claim — it only has ticket data, not order data — so it just takes her word for that specific part.
It ignores the loyalty claim entirely and only discusses the ticket.
Attaching data is specific, not a blanket upgrade to "knows everything." Whatever record you hand it, and nothing else. Order history answers questions about orders. Ticket history answers questions about tickets. Neither one fills in for the other, which is exactly why it's worth knowing what's actually attached before trusting an answer that sounds confident.

Wait — that March ticket

Turn on both toggles above and read the reply again. She wasn't exaggerating to manipulate anyone — she genuinely remembers being let down once before, and this is the second time. That's not something the AI could have known without her support-ticket history attached. It also means there's a promise on record from that March ticket that may never have been kept — and whether the AI actually remembers to check on that next time, instead of you having to bring it up again, is exactly what the next lesson is about.

The part that matters here

"Giving an AI your data" sounds technical, but it's just this: instead of asking it to guess or take someone's word for it, you hand it the actual record. It doesn't make the AI smarter — it makes it accurate, which is a different and more useful thing.

The catch, the same one from lesson 2: none of this happens automatically. Someone has to decide what gets attached, same as someone has to write the rules file. The AI only knows what it's shown.