What A Room Full of Digital Operators Taught Us About AI Literacy

The slides were ready, the agenda was set, but the mood in the room told a different story.

A group of Digital Operations professionals had been pulled from their daily work for an AI workshop. Some had signed up willingly. Others had been told to attend. All of them had an opinion about what they were walking into.

Before we started, we asked one question: what do you think about AI?

We are deliberate about asking it this way. What people think tells you more than what they know or what they have read. The answers that came back were short, direct and honest. Some were a single word. Some were phrases. A few were full sentences carrying the weight of something the person had been sitting with for a while.

They went something like this.

A tool that works without human interference.

Coding and technical. Not for people like me.

Overwhelming.

A threat.

Just an abbreviation.

Difficult to learn and adopt.

Something only large companies can afford.

These are the first impressions of people who run some of the most demanding operational processes in financial services. People who have spent years building expertise in environments where precision is everything and a single error has real consequences.

So why did AI feel so far from their world?

Maybe because the AI conversation in most organisations happens at a strategic level. The people doing the actual work rarely get a version of it that connects to what they do on a Tuesday afternoon. What filters through instead is noise. And noise, over time, becomes either indifference or anxiety. In this room, it had become both.

Three things the room told us

When we looked at the responses, three themes came up.

The technical assumption

Words like "coding", "technical" and "only for large companies" pointed to the same assumption, that AI belongs to developers and engineers, not operations teams.

This is understandable. Most of the public conversation around AI uses language that makes sense to technical people and very little sense to everyone else. When that is the only version of the conversation available, people without a technical background conclude, reasonably, that it has nothing to do with them.

For Digital Operations professionals in financial services, this assumption runs even deeper. Every tool that enters their workflow goes through layers of approval. AI, in that context, sounds like something that would take years to clear and require a dedicated technical team to manage.

What that framing misses is that the most practical AI applications in their world require little to no technical expertise. 

Summarising a document, flagging an anomaly, drafting a routine communication: these do not require an engineering background. They require knowing what problem you are trying to solve. And that is something operations professionals understand better than most.

The fear of replacement

When someone writes "threat" or "work done without human interference" in reference to AI, the instinct is to treat it as a misunderstanding. But these responses came from clarity.

Digital Operations professionals spend their days on work that is high volume, rule bound and repetitive by design. 

They process the same types of documents thousands of times, run the same checks and apply the same logic over and over. And they know that repetitive and rule bound is exactly the kind of work AI handles well. Their concern made sense.

What the "threat" framing misses is the difference between automating a task and replacing a person. The parts of their work that AI can handle are also the parts most likely to cause fatigue and error. Clearing those creates more room for the judgment, context and accountability that operations professionals actually bring to their work.

AI felt too big to approach

Words like "difficult to learn", "just an abbreviation" and "overwhelming" pointed to distance more than resistance.

These professionals hold detailed process knowledge and regulatory awareness that takes years to build. Their capacity to engage with complex subjects is well established. What let them down was how AI had reached them: through headlines about job losses, through technical language that assumed a background they did not have and through communications that described AI's strategic importance without tying it to anything they actually did.

When something is explained at the wrong level and in the wrong language, it becomes background noise. And background noise, for people managing high stakes workloads, gets tuned out. A failure of translation, nothing more.

Three different responses. The same root cause. AI had arrived in their industry without ever properly arriving in their work.

The big shift

By the end of the session, the room felt different.

The language in the feedback had shifted. "Overwhelming" had been replaced by "useful" and "accessible." The idea of AI as a threat had given way to something more practical, an understanding that AI is most useful for the parts of a job that are repetitive, time-consuming and prone to error.

One participant described AI as "a collaborator, a helping hand to reduce task time and a virtual brain." Another wrote that it "can be seen as something that supports people rather than replaces them." Personal conclusions, reached in a room, over a few hours.

That shift came from a combination of things. Simpler language that stripped away the technical jargon. Real examples from industries they recognised. And a consistent reframing of AI as something that works alongside people rather than instead of them.

When you meet people where they are, explain things in terms they already understand and address the concerns they actually have rather than the ones you assume they have, the conversation becomes more meaningful.

The room that walked in carrying noise walked out with a better question: ‘Where does AI fit in what I do?’

What we learned

Running this workshop reinforced something we believe sits at the heart of effective AI literacy work.

You have to listen before you teach.

Walking into a room with slides, frameworks and use cases is the easy part. The harder and more important part is understanding what the people in that room are actually carrying: the assumptions they have built up, the concerns they have never had space to voice and the questions they have been sitting with for months without anyone asking them directly.

That one question we asked before the first slide went up gave us everything we needed to make the session relevant. It told us where to start, what language to use and which assumptions to address first.

AI literacy starts with a conversation. And that conversation starts with listening.