4 AI Prompts That Turn You From an Information Collector Into an Insight Creator

Use this practical toolkit to transform ChatGPT from a lazy research assistant into a powerful intellectual partner.

A week ago, I showed you how I automated my second brain: voice memos automatically transcribed, processed by AI, and organized into my Notion database.

The system works perfectly. I record a thought during my walk, and by the time I get home, it’s already structured and ready to use.

But that’s only half the battle.

I’ve solved the capture problem. Ideas flow into my system without me having to do anything now. Everything’s organized, searchable, and categorized exactly the way I need it.

But the real work starts after the capture.

My Notion database is now full of perfectly organized, AI-processed thoughts. Insights categorized. Ideas structured. Everything searchable and tagged.

And yet most of them just sit there.

Because having information organized in your second brain doesn’t mean you actually know it. Having AI summarize an article doesn’t mean you’ve learned from it. Having your voice memos transcribed and structured doesn’t mean you’ve transformed them into thinking.

The AI automation I built gets ideas into my system. But it doesn’t help me think with them, learn from them, or build frameworks from them.

That’s the gap most people face now. We’ve gotten really good at capturing and organizing information. AI has made that trivially easy.

But we’re treating our second brains like warehouses when we should be treating them like forges. We’re optimizing for storage instead of transformation.

Today’s guest post is from

Eva Keiffenheim MSc

, who’s spent years researching how we actually learn and remember. She’s developed a framework that uses AI not to organize information better, but to transform it into actual knowledge you can use.

Some of her best posts include:

• Why You Forget What You Learn (And How to Fix It in 10 Minutes a Day)

• The 3-Level Prompting Guide to Think and Write Like an Expert (Even If You’re New to AI)

• The Rise of Intellectual Obesity And How to Escape It

This is what most people miss after they build a second brain: the transformation happens when you apply what you learn to your life.

If your second brain is automated but you’re not getting smarter from what you save, this post will change that.

Here’s Eva.

---

Hello, Eva here 👋🏻

I felt a familiar, frustrating paralysis. My perfectly organized ‘second brain’ was failing me. I’d spend more time searching for a note I vaguely remembered than actually thinking with the ideas it contained. I had built a museum for other people’s thoughts.

This isn’t a new problem. Decades ago, philosopher Mortimer Adler argued a clean book signals an unengaged mind. To truly own knowledge, you must argue with it, question it, and leave your intellectual fingerprints all over it.

Adler would be horrified by our digital second brains. But the problem isn’t just that they are unmarked libraries. The problem is that I, like many others, had been aspiring to the wrong thing.

What if the goal isn’t a pristine warehouse, but a messy, productive intellectual forge?

A warehouse stores things. A forge transforms them.

Information is the raw iron ore, brittle and useless on its own. The mind is the forge where, through the heat of focused attention and the hammer blows of critical inquiry, that ore is transformed into tempered steel: novel, flexible, and useful insight.

If you’ve ever felt that frustrating paralysis, overwhelmed by your own beautifully organized system, you’re not a failure. But like me, you’ve fallen for the seductive architectural flaws of the warehouse model.

The act of highlighting, capturing, tagging, and linking creates a satisfying sense of accomplishment. It feels like intellectual work. But as cognitive scientist Robert Bjork’s work on “desirable difficulty” shows, this effortless collection often leads to zero long-term retention. We’ve created a perfect system for what psychologists call the “illusion of competence.” We feel like we know the material in our library, but we’ve merely mastered the art of shelving it.

To break that illusion, we need to stop collecting and start creating. The solution is not a better filing system, but a better set of skills.

Instead of one rigid protocol, here is a flexible toolkit of forging techniques. Each is designed for a specific stage of the thinking process, turning your Claude, ChatGPT, or Gemini from a lazy research assistant into a powerful set of tools for your forge. It provides a set of practical techniques for using AI as the tools of the craft—the bellows, the hammer, the anvil, and the quenching bucket—to shape raw information into durable knowledge.

---

Four Forging Techniques Adler Might Love in 2025

Each technique is designed for a specific stage of the thinking process, turning your AI from a lazy research assistant into a powerful set of tools for your forge.

🔥 Technique 1: Break Your Assumptions (The Bellows)

The Principle: Overcoming functional fixedness. Our thinking gets stuck in familiar ruts. The AI’s non-human associative patterns can break us out by generating novel connections and perspectives.

The Craft: To heat the raw ore of an idea, making it malleable and ready for shaping. To explore the adjacent possible and see your concept in a new light.

Prompt:

#Role: You are a Systems Poet, a master of analogical thinking and conceptual blending.

#Context: My Core Idea: [Insert your core idea, concept, or problem here. Be concise.]

#Task:

##Stage 1: Deconstruct the Core. Analyze my idea and break it down into its 3-5 fundamental components or principles. List them simply as “Core Principles.”

##Stage 2: Refract Through Three Unlikely Worlds. Now, map those Core Principles onto the operating models of the three worlds below. For each world, explain how my idea would be transformed if it obeyed that world’s logic. 

World 1: A Mycelial Network. (Logic: Decentralized connection, resource sharing, symbiotic growth, communication through hidden channels, resilience through redundancy).

World 2: A Tidal Ecosystem. (Logic: Governed by cycles of ebb and flow, periods of intense activity and quiet retreat, adaptation to constant environmental change, value in what is left behind when the tide goes out).

World 3: A Master Artisan’s Workshop. (Logic: Obsessive focus on a single material, deep respect for tools, knowledge passed through practice not theory, value derived from imperfection and the human touch, slow and deliberate creation).

##Stage 3: Generate Actionable Sparks.For each of the three worlds, provide the following output based on your refraction:A “How Might We...” Question: A single, powerful question that reframes the problem based on the world’s logic.A “What If...” Scenario: A tangible, provocative idea or feature that applies the world’s logic to my concept.A New Name: A new, metaphorical name for my idea inspired by that world’s lens.

🔨 Technique 2: The Hammer (Pressure-Testing & Compression)

The Principle: Desirable Difficulty & Retrieval Practice. Forcing yourself to actively defend your ideas against smart criticism is what builds durable, flexible understanding.

The Craft: To beat the impurities (weak assumptions, logical fallacies) out of an idea, giving it shape, strength, and integrity.

AI Prompt:

#context 

I’m going to state a core belief or plan I have. 

#task 

Your task is to act as four distinct personas and attack this belief from your unique perspective. Do not offer solutions or positive reinforcement. 

#guardrails 

Present their critiques as direct, hard-hitting quotes. 

The Cynical CFO: Cares only about ROI, risk, and resources. Views the idea as a potential waste of time and money, and questions its financial viability at every turn.

The Jaded Front-Line Employee: Has seen a dozen grand ideas like this fail before. Focuses only on the practical flaws, the implementation headaches, and how it will make their actual job harder.

The Hyper-Rational Scientist: Cares only about data, evidence, and logical fallacies. Demands empirical proof for every assumption and relentlessly points out unsupported claims. 

The Social Ethicist: Questions the idea’s human impact and fairness. Focuses on who it truly serves, who it might harm, and the underlying power dynamics. 

#My Belief: [State your argument, belief, or plan here]

⚖️ Technique 3: The Anvil (Synthesis & Grounding)

The Principle: Elaboration & Schema-Building. As Barbara Oakley teaches, knowledge becomes sticky when we connect it to what we already know and re-state it in different ways.

The Craft: To provide a solid, stable surface against which to shape the hammered idea, connecting it to first principles and making it memorable.

AI Prompt:

#role 

You are a Knowledge Weaver, an AI that forges durable mental models by connecting new ideas to existing knowledge. 

#context 

My goal is to make my refined insight unforgettable and practical. To do this, you must first understand what I already know. I will provide my insight and the “mental hooks” I already use to understand the world. 

1. My Refined Insight: [Insert your now-stronger insight from the ‘Hammer’ stage.] 

2. My Existing Mental Hooks: [List 1-3 concepts, frameworks, or ideas you know well. Examples: “Supply and Demand,” “The Pareto Principle (80/20 Rule),” “Evolution by Natural Selection,” “A Platform vs. a Pipe business model,” “The concept of ‘compounding interest.’”]

#Your Task: Generate a “Mental Model Briefing” Weave my new insight into my existing knowledge by producing the following four artifacts:

#Guardrails 

The Golden Rule: Distill the insight into a single, memorable, and tweet-able “law” or “rule.”

The Core Analogy (The Weave): Create a powerful analogy that explicitly connects my new insight to one of my existing mental hooks. Explain the connection clearly. This is the most important step.

The Origin Story: Create a short, compelling narrative about how this Golden Rule was discovered. Frame it as either a story of a great success or a costly failure to make the lesson stick.

The Litmus Test: Design a simple diagnostic question or a two-sentence scenario that I can use to quickly test if an action or decision is aligned with the Golden Rule.

💧 Technique 4: The Quench (Application & Solidification)

The Principle: The Testing Effect. Knowledge becomes truly permanent and useful only when it is applied in the real world. Abstract ideas must be made concrete.

The Craft: To rapidly cool and harden the finished insight into a usable tool, making it a permanent part of your intellectual toolkit.

AI Prompt:

#Context 

My Insight (The “Golden Rule”): [Insert the ‘Law’ or ‘Rule’ from the ‘Anvil’ stage.]

The Domain of Application: [Choose ONE: Personal Habit, Creative Project, Team Collaboration, Strategic Decision]

My Risk Tolerance: [Choose ONE: Low (a private observation or thought exercise), Medium (a small change in my behavior that affects only me), High (an action that is visible to others)]

#Your Task: Generate a “Quest Field Guide”

#Output

Based on my inputs, produce a complete field guide for my 48-hour quest. The guide must contain these five sections:

Quest Name: A short, motivating, and memorable name for the mission.

The Prime Directive: A single, ridiculously specific, and simple action I must take. This should be a binary “did it or didn’t do it” task.
Proof of Engagement: The one simple, observable metric that proves I followed the directive. This isn’t about the outcome; it’s about the attempt.

The Pre-Mortem (Anticipate Failure): Identify the single most likely obstacle or excuse that will prevent me from completing the directive. Then, create a one-sentence “If-Then” plan to counter it (e.g., “IF I feel overwhelmed, THEN I will close my laptop and take three deep breaths before proceeding.”).

The Field Notes (Capture the Learning): Provide two simple questions for me to answer after the 48 hours. These questions should be designed to extract the maximum learning, regardless of whether the outcome was a “success” or “failure.”

---

Your Mind is the Workshop

Our digital tools have made it seductively easy to build beautiful, useless warehouses for information. We’ve optimized for collection at the expense of connection.

The framework in this playbook is an antidote. It uses AI not to give you easier answers, but to help you ask harder questions. It invites productive friction back into your thinking. It treats your AI as a powerful but unintelligent set of tools waiting to be picked up by a skilled craftsman: you.

The future of knowledge work won’t belong to those with the largest second brain or the cleverest prompt. It will belong to the intellectual smiths who have mastered the art of the forge—those who know how to transform the endless stream of raw information into the tempered steel of genuine insight.

What idea will you bring to the forge today?

Resources & Further Reading

Adler, M. J. (1940, July 6). How to mark a book. The Saturday Review of Literature, 22, 11–12.

Bjork, E. L., & Bjork, R. A. (2011). Creating desirable difficulties to enhance learning.

Jha, T. (2024). What is the science of learning? (Analysis Paper 63). Centre for Independent Studies.

Newport, C. (2016). Deep Work: Rules for focused success in a distracted world.

My previous guide: How to Use AI Without Cheating Yourself Out of Real Learning. (This piece builds on the cognitive principles outlined there).