The AI readiness divide

From digital divide to capability divide

We spent years worrying about the digital divide — the gap between those with access to computers and the internet and those without. That challenge hasn't gone away, but AI has added a new layer. Access to an AI tool costs almost nothing. The barrier now is capability: the confidence to try, the competence to use it well, and the critical thinking to know when the output is wrong, misleading or just mediocre.

That capability gap is not randomly distributed. It tends to favour people who are already advantaged: those with more education, more discretionary time, more senior roles, more psychological safety to experiment and get things wrong. If organisations ignore it, AI will amplify existing inequalities rather than reduce them.

A spectrum of adopters, sceptics and resistors

Inside any organisation, people's relationship with AI falls across a spectrum. Some embrace it enthusiastically — perhaps too enthusiastically, using it to outsource thinking rather than augment it. Others approach it with healthy caution, asking sensible questions about privacy, accuracy and intellectual property. And some resist it altogether, often for reasons that deserve to be heard rather than dismissed.

• Enthusiastic adopters can move fast and find genuine productivity gains — but may share sensitive data carelessly, accept AI output without scrutiny, or produce what's increasingly called "slop": high-volume, low-quality content that looks polished but lacks substance.
• Thoughtful sceptics ask the right questions about data privacy, intellectual property leakage, accuracy and bias. This caution is a feature, not a bug — organisations need these voices at the table.
• Resistors may be reacting to fear, past poor experiences with technology change, a sense of being left behind, or genuine ethical concern. Their reluctance deserves curiosity, not pressure.

The goal isn't to move everyone to enthusiastic adoption. It's to ensure that no one is left so far behind that they cannot participate meaningfully in a working world that increasingly assumes AI fluency.

Healthy caution versus organisational paralysis

There is a world of difference between thoughtful governance and a blanket ban. Some organisations, faced with uncertainty, have responded by prohibiting the use of AI tools entirely. That approach carries its own risks: employees use them anyway, just covertly; the organisation falls behind its competitors; and the people most likely to find workarounds are, once again, those who were already advantaged.

Healthy caution means clear guidance on what data must never enter an AI system, which tools are approved and why, and how to sense-check output before acting on it. It means proportionate governance — not fear dressed up as policy.

The problem with lazy AI use

One of the risks that doesn't get enough airtime is what happens when people use AI as a replacement for thinking rather than a support for it. Paste in a prompt, accept the output, hit send. The result can look credible while being hollow — or worse, subtly wrong. This matters for inclusion because it can embed bias at scale: AI systems trained on skewed data will reproduce skewed outputs, and if no one is checking critically, that bias gets laundered through a veneer of automation.

Good AI literacy isn't just about knowing how to write a prompt. It's about maintaining the judgement to interrogate what comes back.

What inclusive AI readiness looks like

Closing the AI readiness divide requires action at every level:

• Role-relevant literacy. Generic AI training doesn't land. People need to see how these tools apply to their specific work — and to practise in a context that makes sense to them.
• Safe spaces for experimentation. People learn AI by using it, making mistakes and reflecting. That requires psychological safety — the permission to try without fear of humiliation or punishment.
• Governance without fear. Clear, proportionate policies that address privacy and IP honestly, rather than vague prohibitions that drive use underground.
• Leaders who model responsible use. When senior people talk openly about how they're using AI — what they find useful, what they're cautious about, how they check outputs — it normalises the conversation and gives others permission to engage.
• Accessibility by design. AI tools, training materials and internal guidance should be accessible to people with different learning needs, languages, digital confidence levels and working patterns.

Why this is an inclusion issue

If you care about equity and belonging in your organisation, the AI readiness divide belongs on your agenda. The people most likely to be left behind are often those already navigating other barriers: disabled colleagues who may need specific adjustments to engage with AI tools; people in lower-paid or more operational roles who receive less training investment; those from communities with well-founded historical reasons to distrust technology. Inclusion means ensuring that the AI transition doesn't replicate the inequalities it could, with intention, help to reduce.

Explore how AI connects with belonging and the future of work in the guide AI, Belonging and the Future of Work, or hear these themes explored in depth on the AI & Belonging speaking topic.

Take it further

Read more in the guides library, or hear Joanne explore belonging, culture and the future of work on the Inclusion Bites podcast.