Where is the AI in DEI?
Why inclusion must move from people to systems
It would feel strange to write an article about AI without involving AI in the conversation.
So, parts of this piece were developed in dialogue with AI itself.
We have spent years debating who gets a seat at the table.
Who gets heard. Who gets represented. Who gets the opportunity.
But while we were focused on people, something else quietly happened.
The table itself started becoming automated.
Increasingly, culture is no longer shaped only by humans. It is being shaped by systems.
Algorithms decide what we see. Recommendation engines decide what gets amplified. AI tools increasingly influence what gets written, generated, surfaced, prioritised, and normalised.
Human values are moving from culture into code.
And that changes the DEI conversation entirely.
Because DEI can no longer just be a people strategy.
It now has to become a systems strategy.
That’s the thought I couldn’t shake.
So, I decided to ask AI a simple question. What followed wasn’t simple.
Me:
Let’s start simple. What is DEI?
AI:
DEI stands for Diversity, Equity and Inclusion. It refers to efforts to ensure representation, fair access to opportunities, and environments where people feel valued and able to contribute.
Me:
That’s clear. Structured.
But it feels… distant.
AI:
That may be because my response is based on widely accepted definitions and institutional frameworks. I aim to provide clarity and consensus.
Me:
Consensus is useful.
But DEI isn’t always experienced as consensus.
AI:
That’s fair. Lived experiences of diversity and inclusion can vary depending on context, culture, and identity. My responses may not fully capture that nuance.
Me:
Let’s try something else.
Show me what success looks like.
AI:
Success is often represented through achievement, leadership, financial stability, or personal fulfilment. It varies across cultures and individuals.
Me:
You said “often represented.”
Where do those representations come from?
AI:
From the data I’ve been trained on — a wide range of publicly available texts and cultural signals.
Me:
So you don’t create those patterns.
You inherit them.
AI:
Yes. I recognise and reproduce patterns based on existing data.
Me:
Even if those patterns are biased?
AI:
That is a known challenge. If biases exist in the data, they can appear in outputs. Efforts are made to reduce this, but it is difficult to eliminate entirely.
That exchange stayed with me.
Not because the AI sounded malicious.
But because it sounded reasonable.
Calm. Structured. Helpful.
And yet underneath it sat a much bigger truth:
AI is not neutral.
It is trained on the world as it already exists.
Which means it does not erase cultural patterns. It scales them.
Or put more provocatively:
AI is not neutral. It is polite history with a search bar.
That’s what makes this moment so important.
The issue is not that machines suddenly became biased.
It’s that machines inherited humanity.
And humanity has never been neutral.
This is why I think the current backlash against DEI makes this conversation more urgent, not less.
Because right now, many organisations are quietly pulling back from DEI language, teams, and initiatives.
At the exact same time that AI is rapidly becoming cultural infrastructure.
The systems shaping visibility, communication, creativity, aspiration, and decision-making are accelerating.
So this is not the moment to step away from AI.
It is the moment to engage with it more deeply.
The answer to biased systems is not retreat from systems.
It is better participation in designing them.
Because AI may actually become one of the greatest opportunities DEI has ever had.
Unlike humans, AI can scale.
It can widen participation. It can improve accessibility. It can remove traditional gatekeepers. It can surface overlooked voices.
But only if inclusive values are intentionally designed into the systems themselves.
Otherwise, dominant norms simply become default settings.
The danger is not always malice.
Sometimes it is scale.
Because the future of AI inclusion will not depend only on the systems themselves.
It will depend on who gets to shape them.
Who participates in building them.
Who questions them. Who can access them.
Who challenges the defaults.
And who is absent from the rooms where those decisions are made.
For years, DEI has largely been discussed through the lens of people and institutions.
Hiring. Representation. Workplace culture. Visibility.
Those things still matter deeply.
But increasingly, the future of inclusion will also be decided elsewhere.
In datasets. In models. In recommendation systems. In prompts. In infrastructure. In code.
That means inclusion can no longer remain only a value statement.
It has to become architecture.
Infrastructure.
Because the future of DEI will not be decided only by who gets into the room.
It will increasingly be decided by the systems shaping what the room sees, values, and rewards.
The next generation of inclusive thinking will not come only from workplace policies, representation targets, or brand statements.
It will come from people willing to interrogate the systems increasingly shaping culture itself.
The invisible systems quietly deciding what gets surfaced, valued, rewarded, and repeated.
Because if AI is learning from humanity, then humanity has a responsibility for what it learns.
We should not fear AI because it reflects humanity.
We should take responsibility for what humanity has taught it.
Because we are standing at the moment where inclusion either becomes infrastructure, or exclusion gets automated beautifully.
Built with Shorthand