AI helps with the writing, but you remain in charge: here’s how to avoid a “copy-paste” mindset

In a short period of time, AI has gone from being a tool to the norm, in education, but just as much in organizations. Students can generate a report in seconds. Professionals can use a single prompt to have a plan, analysis, or email created.

And that’s exactly where the challenge lies: how do you ensure that you continue to think for yourself?

Our vision is clear. Embrace AI, but do so responsibly. Not as a shortcut, but as a thinking partner. People remain crucial, especially when it comes to validating information, choosing the right model, and formulating the right task. Those who use AI wisely become faster and better. Those who use AI blindly become dependent and vulnerable.

AI calls for better thinking, not less thinking

AI makes creation easier. That’s why one skill is more important than ever: analytical ability.

It’s not just about knowing how to use a tool, but knowing how to arrive at the right answer. And that applies to everyone, students, teachers, marketers, lawyers, policymakers, engineers, and managers. Wherever AI generates output, the same pitfall arises: the output feels like the truth because it’s written convincingly.

Keep thinking for yourself by taking a step back before sprinting forward. So don’t write right away, first ask yourself:

1. What exactly is the question, and what is the context?
2. What information do I need to support this?
3. How am I going to gather, verify, and select that information?
4. What am I deliberately leaving out because it’s not relevant?
5. What assumptions am I making, and are they accurate?

That process is the difference between generating something and substantiating it. AI can do a lot, but it can’t determine what’s relevant to your context. AI can’t sense which nuances are important in the Netherlands, in your industry, or in your specific case. That remains a job for humans.

From the classroom to the office, everyone faces the same AI challenge

In education, you see it right away. A student submits a report that looks perfect. The question then isn’t just whether they wrote it themselves, but whether they understood it. Understanding lies in the reasoning, not in the sentences.

In organizations, it’s exactly the same, just put in different words. An AI-generated plan may sound logical, but if the assumptions are wrong, if the data is incorrect, or if the context differs, then the plan is worthless or even harmful. This applies to strategy, communication, customer contact, HR, finance, legal, and IT. Everywhere.

That’s why thinking for yourself isn’t just an educational topic. It’s a life skill. AI speeds up the process, but people remain responsible for the choices, the reasoning, and the consequences.

From copy-paste to direction

If you remember just one principle, let it be this: You are the director. AI is merely an assistant.

Taking the lead means you don’t just type a single prompt and accept the answer. Taking the lead means iterating. Verifying. Comparing. Adjusting. And making explicit how you arrived at your conclusion.

A practical method we often use, and one that both students and professionals can apply right away:

1. Conduct in-depth research with AI.
2. Repeat that research using a different model or tool.
3. Overlay the results.
4. Highlight similarities, contradictions, and gaps.
5. Use those gaps to formulate follow-up questions.
6. Only then should you develop your recommendation, plan, or report.

This forces you to think in steps and to treat output as input, not as the truth.

A thinking model that helps you stay sharp: De Bono’s Six Thinking Hats

If you were to choose a single thinking model that helps people evaluate AI output critically and systematically, De Bono’s Six Thinking Hats would be a strong candidate. The model uses six hats, each representing a different way of thinking. You consciously put them on one by one, so you’re not trying to analyze, judge, and be creative all at the same time.

• White hat: facts and data. What do we really know?
• Red hat: feelings and intuition. What does it evoke?
• Black hat: risks and weaknesses. What could go wrong?
• Yellow hat: value and benefits. What are the benefits?
• Green hat: creativity and alternatives. What else is possible?
• Blue hat: process and management. What’s the next step?

The powerful thing is that this also helps you use AI more effectively. You create different prompts for each hat. You force yourself not to try to cram everything into a single answer.

Want to try Bono’s Six Thinking Hats yourself? We’ve created a digital worksheet for you to fill out: Bono’s Six Thinking Hats Worksheet

Example: How to write an advisory report without outsourcing your thinking

Suppose you need to write an advisory report on sustainability in AI. You could be a student, a policy maker, or a team looking to purchase an AI tool. The structure is the same.

Step 1. Blue Hat: plan and scope

Goal: a well-reasoned recommendation that includes options, trade-offs, and measurability.

Questions to define:

• Who is the advisory report for: School, municipality, company.
• What type of AI are we discussing: chatbots, image generation, internal models.
• What criteria determine sustainability: energy, CO2, hardware, data usage, lifespan.

AI prompt example: “Create a research plan with a chapter outline for a recommendation report on sustainability in AI, including sub-questions, data requirements, and validation steps. Context: Dutch organization, practical recommendations.”

Step 2. White hat: gathering facts

Have the AI help you compile definitions and measurement methods. Then you’ll verify them.

AI prompt example: “Provide an overview of measurable factors related to sustainability in AI, such as energy consumption, training versus inference, data centers, and efficiency measures. For each point, include possible types of sources I should consult.”

Your tasks:

• Convert each claim into a verifiable statement.
• Gather primary sources: reports, scientific articles, policy documents.
• Note what is uncertain or context-dependent.

Step 3. Black Hat: risks and bias

Now take a deliberately critical look.

AI prompt example: “Play the role of the critical reviewer. Which claims about green AI are often misleading? Which assumptions are often left unstated? What risks do you see in policy, procurement, and communication?”

Actions:

• Look for counterevidence.
• Check for differences by country, energy mix, and data center.
• Identify greenwashing risks.

Step 4. Yellow Hat: opportunities and value

Now look at the upside.

AI prompt example: “Which measures have the greatest practical impact on more sustainable AI use, sorted by quick wins versus structural changes? Consider model selection, prompt strategy, caching, and governance.”

Actions:

• Link opportunities to the organization’s goals.
• Calculate the impact where possible, and specify assumptions.

Step 5. Red Hat: the human factor

Sustainability is also about behavior and buy-in.

AI prompt example: “Conduct a stakeholder analysis for this topic, identifying typical concerns, motivations, and misunderstandings. Provide suggestions for communication and educational interventions.”

Step 6. Green Hat: alternatives and innovations

Here, you’ll think more creatively.

AI prompt example: “Generate 15 original, feasible interventions to use AI more sustainably in an educational institution, including how to measure and ensure their implementation.”

Actions:

• Select the top 3 to 5 ideas.
• Assess feasibility, costs, impact, and risks.

Step 7. Blue Hat: conclusion and recommendation framework

You then pull everything together into a clear, explainable recommendation.

Final structure:

1. Executive summary
2. Problem and context
3. Research approach and validation
4. Analysis and findings
5. Options and considerations
6. Recommendations and roadmap
7. Measurement plan and governance
8. Appendices with sources and assumptions

The hardest step: validation. How do you know if it’s correct?

This is where things often go wrong. AI can write convincingly, even when it’s wrong. That’s why validation should be an explicit part of the process, not an afterthought.

A practical validation approach that anyone can learn:

• Apply a source hierarchy: give priority to primary sources such as scientific publications, official statistics, policy documents, and direct reports from institutions. Use blogs and summaries only as a guide.
• Check claim by claim: Don’t let AI create a single narrative that you then have to unravel. Break down the claims and verify them individually.
• Triangulation: verify key claims using at least two independent sources. Don’t rely on two articles that simply copy the same press release.
• Context check: Always ask whether this applies to the Netherlands, to this sector, to this year, or to this target audience. Much of the output is generic or biased toward the U.S.
• Require a counterargument: for every conclusion, present a serious counterargument and explain why you’re not accepting it.
• Transparency in your work: note any assumptions, uncertainties, and data points you were unable to verify. That’s not a weakness, it’s professional.

This is exactly how you force yourself to keep thinking. You make your reasoning transparent.

AI as both a threat and a lever

Yes, AI is changing jobs. Some tasks will disappear or become less significant. That’s realistic. But there’s another side to it, too.

AI is a lever. Those who learn to use AI effectively can adapt more quickly, provide better justification, and deliver greater value. And that creates a new kind of distinction. It’s not about who can generate the most polished text, but who can make the best decisions based on sound reasoning.

This is especially relevant for students and young professionals. They’re entering a job market where AI is the norm. Their competitive edge doesn’t lie in knowing a single tool, but in being able to think in steps, validate their work, and explain why something makes sense.

Banning it is just treating the symptoms; thinking for yourself is the real skill

The question isn’t whether people will use AI. The question is whether they’ll learn how to remain responsible for their own decisions.

When education and organizations embrace AI with frameworks, thinking models, and validation routines, they empower people, not make them more dependent. That way, people learn to collaborate with technology without losing their own judgment.

“Bullshit in, bullshit out” remains true. The difference is that we now have the opportunity to teach everyone, students and professionals alike, the opposite. Critical thinking as the foundation. AI as an accelerator. And humans as the ones ultimately responsible.