IQ is supposed to be an indicator of intellectual intelligence, how well you understand and solve problems.
EQ is supposed to be an indicator of emotional intelligence, how well you manage emotions and social interactions.

Both matter in the workplace, where you need to think clearly, work effectively, and get along with your colleagues.

But with the rapid arrival of artificial intelligence in the workplace, we might need another measure for thinking and behaviour — AI-Q.

Think of AI-Q as the human wisdom required to handle the artificially intelligent colleague who has suddenly arrived in the office, thanks to the rise of generative AI tools such as ChatGPT.

It’s the skills humans need to direct AI to be a genuinely helpful assistant, the way we will collaborate with AI as a colleague, or even the way we will manage upwards if (when) AI is the boss.

So what does developing AI-Q entail?

When Microsoft considered the question of whether AI could fix work earlier this year, it argued humans in the workplace were sufficiently unhappy with the work they did that they would embrace AI as a colleague.

The “digital debt” of data, emails, meetings, and notifications “has outpaced humans’ ability to process it all,” it said — a bit rich from the company that brought us Excel, Outlook and Microsoft Teams.

Two out of three workers reported struggling with having enough time and energy to do their job, the report found, and those people were 3.5 times more likely to struggle with innovation and strategic thinking.

The advance of AI, it said, would lead to a new “AI-employee alliance,” and humans would happily hand off tasks to their new robot roomies.

“While 49% of people say they’re worried AI will replace their jobs, even more—70%—would delegate as much work as possible to AI to lessen their workload.”

Microsoft: Will AI Fix Work

This rose-coloured future obscures a genuine challenge for organisations.

People are used to thinking of technology as tools — Microsoft Word hasn’t talked back since the days of Clippy — but AI needs to be conceptualised as a co-worker to get the best out of its potential.

Let me spell that out.

If you think of and treat your AI as an intern, you get intern-level responses. That’s often what you see when people are still asking ChatGPT to ‘write me this’ or ‘create me that’ with no other instruction or context.

But AI can be any role you need.

So if you ask it to reflect the thinking of a worker with a decade of experience under its belt, that’s what you get instead.

Give it a goal, role, constraints and examples, and you get dramatically better outputs.

Tell AI that it is a specialist in a particular field, and you will get an answer through that lens.

Require AI to revise its responses to be smarter, more cogent, more strategic, more persuasive, then it will deliver that as well.

A key step for organisations, then, is to help introduce the idea of AI personas, so human workers know who they are really speaking to — the intern, the crusty but experienced colleague, the industry expert, the domain specialist — when they collaborate with it on work.

The second challenge is to manage the discomfort this will cause.

Humans work hard to be good at the things they do, and no one wants to be usurped by technology.

It can be tempting to look at the speed at which ChatGPT churns out text or Midjourney creates images and assume talents like writing or designing are obsolete.

“But just as a chisel is more effective in the hands of a master craftsman than in that of a hobbyist, AI is also more effective when used by an expert in a particular knowledge domain”.

Organisations looking to embed generative AI might start by helping humans identify their own personal weaknesses and strengths, with some guidance around how AI can mitigate the first and build on the latter.

Personally, that means using ChatGPT to fill in gaps in my skills.

I often ask it to figure out what formula I need to drop into Excel to tidy up data (“How do I do a VLOOKUP again?” “What is the right formula to split text?”)

I ask it for timings for a workshop so I don’t have to waste time on a plan, and I use an image generator rather than a sketch of stick people to show a designer what’s inside my head.

And for tasks where I have the requisite skills, I can use AI to push beyond what I’d normally deliver.

I can turn one piece of work into four different formats, incorporate additional viewpoints, or break a single idea into many.

Microsoft’s guidance here is for organisations to help their employees develop what it calls AI aptitude — AI-Q by another name.

That includes helping people understand where and when AI can be used in work and practice the instructions and requests that are going to help them get the best out of any machine-human collaboration.

Just nine months into the generative AI revolution, we are just at the edges of the workplace change that will soon be felt.

The time to help your humans build the intelligence needed to adapt is now.