Looking in the back of the book? Why critical thinking must come before AI

Last week – like many parents across Australia – I found myself trying to unsuccessfully balance work and school holidays.

Surrounded by heated negotiations over Fortnite and the occasional airborne cloud of flour, baking with my two sons quickly turned into a scene of disorder – ingredients flying, laughter and very little resembling the neat process in the recipe.

The cake mixture flying out from the egg beater got me thinking what we were doing wasn’t about getting the perfect cupcake, it was about learning how to learn and it’s a lesson I think many business leaders need to hear particularly when it comes to artificial intelligence (AI).

Increasingly, AI is becoming the grown up version of looking in the back of the book for answers.

Just like in school, when you skip the hard thinking and go straight to the solution, you miss the point entirely.

AI is not a shortcut, it’s a test

Many of the organisations I work with come to us with a version of the same problem: ‘We want to implement AI, but we’re not sure what to do with it.’

This is often followed by tactical wish list items such as ‘automate X, summarise Y, save time on Z’ – without much reflection on where AI could be of benefit in the context of their business model.

Too many companies treat AI like a shortcut to value.

But AI isn’t magic. It doesn’t come with strategy baked in. It doesn’t replace thinking. In fact, what it often does – especially in the early stages of adoption – is reveal just how little critical thinking is happening in the first place.

This is where the “back of the book” analogy becomes useful.

When a student jumps to the answer without understanding the steps, they might get an A but they won’t grasp the concept, they can’t apply it next time and they’ll struggle the moment the variables change.

Business leaders are no different. When we let AI give us answers without understanding the assumptions, the logic, or the trade offs behind them, we risk trying to build skyscrapers on sand.

Kids learn by doing, so should leaders

We coach our kids to work through problems: Show your working. Try it yourself first. Don’t be afraid to get it wrong. In contrast, some executives are skipping straight to the AI output and presenting it as truth.

That’s dangerous not because the technology has inherent flaws (e.g hallucinations and bias), but because the human oversight is missing.

What’s needed now isn’t just access to AI. It’s the discipline and critical thinking to think before we prompt. To understand the problem and process before we automate, to interrogate before we accept.

Many executive teams we work with need to commence with uplifting their understanding of AI both in terms of capability and where the technology applies to their critical business objectives. This involves:

• Starting with executive alignment and AI literacy
• Identifying the highest value opportunities, the critical workflows where AI can shift the dial
• Conducting technology feasibility of enhancing these high value opportunities
• Identifying the value and sequence of activities
• Building a fit for purpose roadmap (people, process, technology data, governance)

Because when leaders understand their objectives and current state process, AI becomes an amplifier; they can think big, start small and scale fast.

Critical thinking is the real edge

The most successful AI implementations don’t start with ChatGPT or Copilot, they start with robust conversations about workflows, decisions and capability gaps. They are led by truth tellers in your business who are willing to ask questions that don’t have neat answers.

Ultimately, the organisations that win in this next era won’t be the ones that use AI the most. They’ll be the ones that are clear on their strategy and who they serve. They will extend their advantage by implementing AI in heart of their operations aligned with the organisation’s core competence.

So yes, AI is like the answers in the back of the book. But before you look back there, make sure you’ve done your thinking. Your business and your leadership depend on it.

Lindsey Hershman is the Managing Director of AI Access, a full-service AI firm with expertise in AI training, strategy, governance, custom AI development, testing and adoption. AI Access assists organisations to ‘avoid the hype, stay in control and realise the value’ by linking and integrating strategy, technology and people.