A visual guide

Smarter software, or AI colleagues?

Software suppliers are adding AI to every system. It helps the person who is logged in. An AI colleague does a different job: it carries the work between the systems you already have and brings it back for you to check.

In brief

This CogniScale visual guide compares two ways of bringing AI into an organisation. AI built into a software product helps the person using that product. An AI colleague carries the work between the systems a person already uses, through that person’s own access. The page ends with three questions to ask a software supplier before an AI upgrade.

Picture a kitchen

A smarter oven does not cook dinner.

Smart appliances
Fridgesmart
Hobsmart
Ovensmart
Table

The fridge knows what is inside it. The oven suggests a temperature. Each is a little better than the one it replaced. Dinner still does not make itself: someone has to take the food from the fridge to the hob to the oven to the table.

A cook
Fridge
Hob
Oven
Table

You say what you would like for dinner. The cook works in the kitchen you already have, with the appliances you already own, and brings you the plate to taste before it goes out.

Buying each system again with AI inside it is the smart appliance. An AI colleague is the cook. The systems stay where they are. What changes is who carries the work between them.

The same idea at work

Who carries the work between your systems?

Think about the person who moves between the CRM, the finance system, spreadsheets and email. The systems are the same in both pictures.

AI inside each systemEach system gets an assistant.CRM+ AIFinance+ AISpreadsheets+ AIEmail+ AIYouNothing joins them togetherexcept you
Each assistant helps inside its own product. You still carry the work from one system to the next.
An AI colleague across your systemsOne colleague works across them all.CRMFinanceSpreadsheetsEmailgives the jobchecks the resultYouAI colleague
The colleague works through your own access to the systems you already have. You give the job and check the result.
Illustrative.
One job, two ways

The hour after a sales meeting.

Five tasks, none of them difficult. The question is who does them.

With AI inside each systemYou do all five tasks. An assistant helps with each.
1Write up the notesYou
2Update the customer recordYou
3Hold the dates in the diaryYou
4Draft the follow-up emailYou
5Start the proposalYou
With an AI colleagueYou give the job and check the result. The colleague does the five tasks.
You give the jobYou read it and press send
1Write up the notesAI colleague
2Update the customer recordAI colleague
3Hold the dates in the diaryAI colleague
4Draft the follow-up emailAI colleague
5Start the proposalAI colleague

After the second meeting the colleague starts from what it recorded after the first. Nothing goes to the client until you have read it and pressed send.

Illustrative.
Side by side

An assistant answers. An agent is given a job.

Where it works
AI inside each systemInside one product.
An AI colleague across your systemsAcross the systems you already use, through your own access.
How the work starts
AI inside each systemYou open the system and ask.
An AI colleague across your systemsYou give it a job.
How far it goes
AI inside each systemIt helps with the task in front of you.
An AI colleague across your systemsIt works through the steps, checks its own result and tries another way when something does not match.
What it keeps track of
AI inside each systemThe task you have open in that product.
An AI colleague across your systemsWhere the job has got to, so it can start from there next time.
Who joins the work together
AI inside each systemYou do, moving from one system to the next.
An AI colleague across your systemsThe colleague does. It comes back to you when a decision is needed.
What you buy
AI inside each systemA newer version of each system.
An AI colleague across your systemsA way for AI colleagues to work with the systems you have.
Who stays in charge
AI inside each systemYou.
An AI colleague across your systemsYou. By default, nothing is sent or published without your approval.

Both have their place. AI inside a product makes that product easier to use, and some of those upgrades will be worthwhile. It does not carry the work to the next system.

Before you buy an AI upgrade

Three questions for any supplier.

1Does your AI work only inside your product, or can it act in our other systems too?
2Can an outside agent read and update records in your system, through a documented connection and under our permissions?
3When your AI finishes a step, who carries the work to the next system: it, or my member of staff?

There is an exception. Some older systems offer no way in at all, and replacing those may be the right call. The point is to find out before you buy.

Your own systems

Start with the systems you already have.

The AI Opportunity Report combines leadership priorities, team experience and systems constraints to identify where AI colleagues could create value first.

Read Tim Bond's article