The Problem: AI Tool Confusion
In today's rapidly evolving technological landscape, artificial intelligence tools have become increasingly accessible to the average person. From ChatGPT to Midjourney, Copilot to Claude, we're surrounded by AI assistants promising to make our lives easier and more productive. But with this proliferation comes a critical question: When should we actually use these AI tools?
Many of us find ourselves in one of two problematic camps:
Over-reliance: Using AI for everything, even tasks where developing personal skills would be more valuable in the long run
Under-utilization: Avoiding AI tools altogether due to confusion, fear, or uncertainty about their appropriate applications
What we need is a simple, practical framework to guide our decisions about when to leverage AI and when to rely on our own capabilities.
Introducing the FLD Framework
The FLD (Fix, Learn, Do) Framework provides a straightforward approach to deciding when AI is appropriate for your tasks. It centers around three fundamental questions you should ask yourself before turning to an AI assistant:
Fix (F): Do you know how to fix that if it goes wrong?
Learn (L): Do you want to learn how this works while doing it?
Do (D): Do you just need to complete the job regardless of how it works?
These three questions help clarify your goals and determine whether AI is an appropriate tool for your current situation.
Understanding the Framework
Let's break down how the FLD Framework guides your decision-making:
Fix (F)Learn (L)Do (D)Should You Use AI?ExplanationYesNoYes✅ YesYou need to ensure the task is done correctly and be able to fix issues, but don't need to understand the underlying processNoYesNo✅ YesYou're primarily interested in learning, with no pressure to complete the task efficientlyNoNoYes❌ NoWhen you don't need to fix potential issues or learn the process, but just need something done, AI dependency can create risksYesYesYes⚠️ Limited useUse AI as a collaborative tool, not a replacement for your workYesNoNo❓ SituationalConsider if AI can serve as a reliable safety netNoYesYes⚠️ Limited useUse AI for guidance, but do the work yourself to learnYesYesNo✅ YesIdeal for learning with a safety netNoNoNo❓ SituationalReconsider if this task is necessary at all
Practical Examples
Scenario 1: Writing Code for a Critical Production System
Fix: Yes (you'll need to debug and maintain it)
Learn: Yes (understanding the code is essential)
Do: Yes (it needs to be completed)
Recommendation: Limited AI use. Use AI for suggestions and to learn patterns, but write critical sections yourself.
Scenario 2: Creating a Family Birthday Card
Fix: No (minor imperfections are acceptable)
Learn: No (the process isn't educational)
Do: Yes (you just need a nice card)
Recommendation: No AI. The personal touch matters more than efficiency.
Scenario 3: Learning a New Programming Language
Fix: No (it's for learning, not production)
Learn: Yes (that's the whole point)
Do: No (completion isn't the priority)
Recommendation: Yes to AI. AI can provide examples, explain concepts, and help you understand patterns.
Scenario 4: Writing a Business Proposal
Fix: Yes (errors could be costly)
Learn: No (you don't need to learn proposal writing techniques)
Do: Yes (you need it completed efficiently)
Recommendation: Yes to AI. Have AI draft sections and focus your time on reviewing and customizing.
The Core Decision Matrix
At its simplest, the FLD Framework can be distilled to this essential decision matrix:
Fix (F)Learn (L)Do (D)Use AI?YesNoYes✅ YesNoYesNo✅ YesNoNoYes❌ No
This simplified matrix captures the most important combinations that determine whether AI is an appropriate tool.
Why "No-No-Yes" Means "No AI"
The combination of "No" to fixing, "No" to learning, but "Yes" to just getting something done represents the most dangerous use case for AI. Why?
When you don't care about learning how something works and don't plan to fix it if something goes wrong, but still need it done, you're creating a dependency without accountability. This situation:
Builds technical debt
Creates potential points of failure with no contingency
Develops dependency without understanding
Reduces your agency and capability over time
In these cases, you're better off either:
Hiring a professional who can take responsibility
Taking the time to learn enough to be accountable
Reconsidering if the task is truly necessary
Conclusion: Intentional AI Usage
The FLD Framework isn't about restricting AI use—it's about making intentional choices regarding when and how we incorporate these powerful tools into our lives. By asking yourself these three simple questions before reaching for an AI assistant, you can:
Develop skills that matter to you
Leverage AI where it truly adds value
Maintain autonomy and accountability
Build a healthier relationship with technology
As AI capabilities continue to advance, frameworks like FLD become increasingly important. They help us navigate the balance between technological assistance and human development, ensuring we use AI as a tool for empowerment rather than a crutch that diminishes our capabilities.
Remember: The best use of AI isn't about maximizing how often you use it, but about maximizing the value it brings to your life and work.
Your Turn
Consider a task you're planning to use AI for today. Apply the FLD Framework:
Do you need to fix it if something goes wrong?
Do you want to learn how it works?
Do you just need to get it done?
Your answers will guide you to the right decision—not just for efficiency today, but for your capability and growth tomorrow.