AI Readiness: What about the humans?

02/10/2026

AI readiness:  What about the humans?

Short on time? Jump to the six questions every leader should ask about their team’s Human Readiness.

Picture the scene. Maybe even put yourself in it.

It’s the meeting about AI that’s been sitting in the calendar for a month. The demonstration of what the technology can do has been impressive, jaw-dropping in places.

Someone is asking all the right governance questions. Somebody else has calculated the projected time saving post-implementation. Around the table, there is a slow Mexican wave of nodding agreement that the team or organisation needs to move.

Then the discussion reaches a familiar point.

Who can authorise the next step?

Everyone looks at you.

You catch the eye of someone who has barely spoken all meeting and suddenly realise their expertise is probably the most affected by what you have just been discussing. You suspect the questions about the proposed “savings” and “efficiency” may surface later, in smaller conversations where you are not present.

Two functions are exploring remarkably similar ideas, although neither knows it yet.

So you leave with a promising plan and the vague unease that making it happen will depend on different parts of the team working well together, and therefore, pretty heavily on you too.

Perhaps there is another thought you are less inclined to say out loud:

Am I actually ready to lead this well myself?

The technology on show in the meeting is genuinely cutting edge.

Almost everything else that happened is very much business as usual.

And not in a good way.

People problems do not sit alongside AI adoption 

Technology, data, governance, skills and training all deserve serious attention.

AI also enters an existing set of relationships: who gets heard, what can be challenged, how decisions really get made and where responsibility eventually lands.

A team can appear to be working perfectly well because a handful of capable people may compensate for what is difficult between everyone else. They chase, smooth things over, make the decision, join the dots and generally prevent the wheels from coming off. It’s tempting to think, yep…and?

The problem is, as change accelerates and familiar assumptions become less reliable, the workarounds get much harder to sustain.

These are not separate “people problems” sitting alongside AI adoption, they’re part of whether adoption works, or whether it sinks like a bad sourdough in a cupboard.

And you are inside this change too.

You are expected to protect today’s performance while preparing people for a future you can’t yet fully describe. You may well be excited by what AI makes possible and uneasy about what it could mean at exactly the same time.  That is a not uncommon prevailing mood in SLTs and boardrooms up and down the country, and you’re not alone. 

You want people to experiment, and you also know that if something significant goes wrong the accountability will still wind its merry way back to you.

You may worry about moving too slowly, backing the wrong horse or discovering that the capable team around you needs more support than you currently have left to give.

Then there is the question that tends not to make it into strategy discussions:

How much uncertainty can I admit to before people start wondering whether I am the right person to lead this?

Good news…

You don’t have to perform certainty to offer direction.  

You do need a way of working through uncertainty with your team though, rather than becoming the place where everyone else’s uncertainty accumulates.

When every AI decision comes back to you

A small experiment needs a decision. Someone sends it upwards “just to be safe”. You answer because it is quicker than explaining why they could have decided themselves, or coaching them to enable their own decision-making, and so the work moves forward.

The pattern stays exactly where it was.

It’s easy to see how this happens, because you remain accountable for the consequences, and delegating judgement can feel uncomfortably like accepting risk without retaining control.

There can also be reassurance in being needed, even when it exhausts you. 

Over time, though, a team can become extremely skilled at identifying what needs your approval and much less practised at deciding what doesn’t. Your frustration with their dependence may sit alongside habits that help maintain it.

That is an uncomfortable thought, especially when some of those habits probably helped you succeed in the first place.

Human readiness for AI means being clear about what people can try, what they can decide and when new evidence genuinely needs to bring something back to you.

It also means noticing the difference between a decision that requires your authority and one you would simply feel more comfortable making yourself.

To put it simply, the phrase “use your judgement” only really works when people experience you actually letting them.

What the AI-related silence really contains

Let’s return to the colleague who barely spoke in the meeting. They may have spotted a genuine flaw in the proposal. They may also be wondering what happens to their standing when something they spent years mastering suddenly becomes way easier for other people (without the mastery) to do.

Both can be true at the same time.

Treating hesitation as resistance misses something really important: People are being asked to question their expertise while remaining confident in their value. That’s a tough one. Especially so for people with an introverted personality preference.

Leaders face their own version of it when authority has long rested on being the person who has all the answers.

It is tempting to reassure quickly.

• This will free everyone up for more interesting work.

• • That is a much easier sentence than:

• We do not yet know exactly what this means for your role.

To name a few…

Reassurance that overshoots what you actually know can make the next honest conversation much harder than it needs to be.

And honesty can be uncomfortable for the leader too.

You may genuinely want candour and still feel your stomach sink when someone says the thing that complicates the plan you have spent a month getting everyone behind.

That is where candour gets real. The question isn’t simply whether you invite people to speak, it’s what happens when they interrupt the momentum with something you would rather not hear – not now anyway. 

Staying curious means being prepared to examine it properly before deciding what happens next, yet sometimes it can inadvertently get conflated with agreement.

The same applies when nobody seems concerned.

An AI-generated analysis lands looking polished, persuasive and extremely finished. You’re rightly enthusiastic. Then, someone notices an assumption that doesn’t really fit your business, only now, challenging the output feels slightly like challenging you, and let’s face it, at this point, they may think twice about going there. 

AI is very good at making weak thinking look respectable.  (That’s an understatement actually!)

A fluent answer still needs a human willing to test it.

Why the AI pilot never becomes the work

Three months later, the experiment is still “really promising”.

The demos keep improving, yet the working week looks suspiciously similar.

The original process remains because nobody has agreed what can stop yet.  The new approach belongs unofficially to the enthusiast who introduced it. Everyone supports the idea. Nobody actually owns the change.

This is where ownership becomes more demanding than simply giving people permission to experiment. Someone needs the authority, time and support to change the workflow, judge the results and decide whether to continue, adapt or stop. They also need to know that an agreed experiment isn’t  suddenly going to be treated as reckless just because it didn’t turn out quite as hoped.

For leaders, there is an obvious temptation to give the job to the person who always delivers.

Why? Because it gets done – Responsibility concentrates even further in the same reliable few.

Before asking why people haven’t taken ownership, try asking yourself what they’ve genuinely been given the autonomy  – and capacity – to change.

Have you really handed over ownership? Or have you simply handed over the work while keeping the meaningful decisions yourself?

A faster AI version of the same doesn’t move you forward

Somewhere else in the building, sales is using AI to produce proposals faster than ever, and delivery uses it to summarise handovers.

Both functions can tangibly demonstrate progress.

Neither has addressed why the handover routinely loses information the other team needs.

Problem officially accelerated!

Some of the biggest opportunities created by AI will undoubtedly sit between functions rather than neatly inside them. That requires people to question whether the existing process deserves speeding up in the first place – it also means dealing with territory.

One function owns the relationships, another owns the data. Somebody controls the process, someone influential designed the thing everyone is now wondering whether they still need.

Perhaps that someone was you?

A leadership team can ask people to collaborate while continuing to reward them primarily for protecting their own targets, budgets and territory. People generally work out very quickly which expectation really matters. 

The practical test is whether functions can change something together when the effort and the benefit do not fall neatly in the same place.

You can be trusted without pretending to be certain 

Trust runs underneath all of this, it’s the footing of any brilliant  team. 

Trust that questions will be heard, responsibilities will be honoured, people can challenge without being punished for it and an intelligent mistake will create learning rather than a sudden outbreak of approvals and nervous supervision.

Your team is already learning how to respond to uncertainty by watching what you do with it.

• What happens when somebody disagrees with you?
• When an experiment fails?
• When a decision could be made without you?
• When one function needs to give something up for the wider organisation to gain?

None of this means every team habit is yours to own. Teams accumulate patterns through history, incentives, personalities and pressure, and some of the habits that now get in the way probably once helped the organisation get to where it is today. 

The human side of AI readiness

The hard part of leading AI change is not having all the answers. It is creating enough confidence for people to move while being honest that nobody has them all.

That is why our Human Readiness Assessment looks at four conditions that matter increasingly as AI accelerates change:

Decision-making. 

Candour and conversation. 

Ownership. 

Collaboration and creativity.

And perhaps the most useful question for a leader is not simply: How ready are my people?

It is:

What have we taught them about how decisions get made, how safe it is to challenge, what happens when somebody gets something wrong and where responsibility really sits?

The better the humans work together, the more value they will get from AI.

So alongside the investment in technology, governance, skills and training, there is another question worth asking:

How human-ready is the team expected to make AI work -and what is my leadership teaching them about how to handle what comes next?

 

 

Six questions to ask about your team’s Human Readiness

AI readiness is not only about what technology your organisation can use.

It is also about what happens between the humans when the answers are moving, the stakes feel personal and nobody has a perfect map.

So before asking whether your team is ready for AI, ask:

1. Can we be clear without pretending certainty?

Does the team know enough about the direction, boundaries and next step to move?

And can you say “we don’t know yet” without feeling you have failed to lead?

Clarity does not require you to predict the future. It means being honest about what is known, what is still changing and what happens next.

2. Can we listen without becoming stuck?

Can people raise genuine concerns, question the direction or say what they think could be lost?

And once those things are on the table, can the team examine them properly and still make a decision?

Candour matters. So does movement.

3. Can we experiment without losing judgement?

Do people have enough freedom to try something, learn and adapt without sending every decision upwards?

And when an experiment does not work, does the response create learning or suddenly create three new approval stages?

Experimentation only works when autonomy and accountability live together.

4. Can we change the work without carelessly diminishing people?

What happens when AI touches something somebody has spent years becoming excellent at?

Are you treating hesitation as resistance, or getting curious about what may be underneath it: expertise, identity, status, pride or uncertainty about where their value sits next?

Changing a task is sometimes easy.

Changing what that task means to a person is not.

5. Are we using AI to strengthen human thinking, or replace it?

Does a polished AI answer get examined, challenged and tested?

Or does “the AI says…” quietly end the conversation?

The technology can help people think further and faster. Somebody still needs to question the assumptions, understand the context and stand behind the judgement.

6. Is ownership genuinely shared, or has responsibility simply become vague?

Who can make the decision?

Who changes the workflow?

Who decides whether an experiment continues, adapts or stops?

And when everyone says “AI is everyone’s responsibility”, does that actually mean anything different happens?

Shared ownership should spread responsibility, not make it disappear.

 

Start with a clearer view

Our free Human Readiness Assessment explores decision-making, candour and conversation, ownership, and collaboration and creativity.

Answer 15 questions and get personalised insights into where your team is strong, where change may snag and what deserves attention next.

 

Take the Human Readiness Assessment