Every month we have an open webinar with the great and good of the accounting and finance in world in business. In house finance teams trying the best to do the day job and benefit from the technological evolution which is happening around them, and around us all!

This month we talked about cash, treasury and the robots!

What I found most interesting was the large number of attendees who would not use AI for cash decision, 72% in total. But why?

Are we already automation cash decisions with cash pooling, and we already using AI to forecast positions, and we already asking AI to advise on cash strategy… so why the reservation? 

If you’ve ever found yourself nodding along in meetings about AI strategy without really knowing which flavour of AI is being served, it would appear you are not alone. 

Because here’s the truth… Not all AI is created equal. And not many of us get it!

And I understand why.

Today, we’re drowning in headlines about Artificial Intelligence, but beneath the buzzwords are very real, very different tools, each with their own strengths, weaknesses and use cases. And knowing the difference can help you lead the change, not follow the hype.

So let’s break it down. 

Three types. 

Three different superpowers.

Machine Learning (ML): The Number Cruncher

ML is the quiet workhorse of AI. It learns from historical data and spots patterns to make predictions.

Think - forecasting revenue, predicting churn, spotting fraud.

Strength - numbers; time series, regressions, probabilities.

Use it when you have loads of clean, structured data and need repeatable insights.

ML is great at what we used to do. It tells you what might happen based on what has happened. It needs training data, a statistical model, and off you go!

It’s not the shiny new toy anymore. In fact, it’s a well-worn toolkit; reliable, solid, serious about numbers.

Generative AI (GENai): The Wordsmith

This is the ChatGPTs and Copilots of the world. GenAI doesn’t just analyse data, it can create it, new content based on old.

Think - drafting reports, writing emails, summarising documents.

Strength - Words; language, context, human-like output.

Use it when you want to save time on comms, or translate complex ideas into something people actually understand.

In finance? It can auto-generate board summaries from your dashboards, write your month-end commentary, or help you prep for stakeholder meetings — in seconds.

This is where automation meets creativity and personal productivity.

Agentic AI: The Self-Starter

Now we’re talking sci-fi or what was sci-fi until recently. Agentic AI doesn’t just generate or analyse. It acts. It decides. It’s goal-driven.

Think - Job done!, “Book this meeting, send the deck, follow up with Sarah”.

Strength - Autonomy; multi-step tasks, decisions, operations.

Use it when you want to delegate whole processes, not just parts.

In finance, that might mean a bot that reconciles accounts, flags anomalies, and emails the relevant team without you touching a thing.

This is automation, upgraded. Not just doing tasks, owning them. The CFO’s digital chief of staff.

So… Which One Should You Use?

That depends on the job. And let’s be honest, the job is changing fast.

So don’t chase tech for tech’s sake. Look at what your organisations needs and then decide how to meet it.  

In the GENCFO Talks on cash and treasury, we were in fact talking about all 3 AI’s; ML, GENai and Agentic AI.

So don’t get caught up in the AI soup. 

My advice is, learn enough to be inspired and have an opinion on the tech, but they start with your business challenge, your capability gaps, your goals. 

Then ask: Which AI actually solves this?

Because yes, AI. But let’s be bold enough to ask, Which AI?

Want to join in these conversations in real time? Take a look at our membership options and be part of the discussion!

Become A Member