I Am Not a Robot My Year Using AI to Do (Almost) Everything

In one sentence

To master AI, you must stop treating it as a search engine and start treating it as a tireless, slightly eccentric intern that requires clear instructions, constant feedback, and a human "editor-in-chief" to provide the final spark of judgment.

Why read this (and who it's for)

This book is for the "AI-curious" who are overwhelmed by the sheer volume of tools and hype. It solves the problem of how to actually integrate AI into a daily workflow without losing your own voice or sanity. If you’ve tried to use ChatGPT and felt frustrated because it gave you generic, "hallucinated," or useless answers, this book explains why that happened and how to fix it. After reading, you will be able to architect complex multi-step prompts, use AI as a creative sparring partner, and identify exactly which parts of your life are "AI-ready" versus which require a human touch.

The big idea

The central thesis of Stern’s year-long experiment is that the primary barrier to AI proficiency isn't the complexity of the software—it's the shift in human cognition required to use it. Most people approach AI with a "Google mindset": they type a query and expect a perfect result. However, AI thrives on a "Collaborative Mindset."

Stern argues that AI is a force multiplier for intent. If your intent is vague, the output will be mediocre. If your intent is sharp, the output can be extraordinary. The book illustrates that the real power of AI lies in its ability to handle the "drudge work" of synthesis, brainstorming, and drafting, allowing the human to remain the sole arbiter of taste, ethics, and final execution. It is a journey from automation (replacing a task) to augmentation (enhancing a capability).

Key ideas

The Prompt as a Brief

What it is — A prompt is not a search query; it is a creative brief. It must include context, persona, task, and constraints. Why it matters — When you treat a prompt like a Google search, you get generic "average" data. When you treat it like a job description for a talented freelancer, you get bespoke, high-quality results. In the book — Stern demonstrates how providing a "persona" (e.g., "Act as a world-class travel agent who specializes in budget-friendly hidden gems") fundamentally changes the depth and relevance of the output compared to a simple "Where should I go in Italy?"

The Iterative Loop

What it is — The first response from an AI is rarely the final product; it is a "first draft" meant to be critiqued, refined, and expanded. Why it matters — Expecting perfection on the first try leads to frustration. Embracing the "conversation" allows you to guide the AI toward the specific nuance you need. In the book — Stern shares her experience of "chaining" prompts—taking a piece of information the AI gave her, asking it to expand on a specific bullet point, and then asking it to rewrite that bullet point in a more professional tone.

The Friction of Automation

What it is — Sometimes, using AI actually takes more time than doing the task yourself. Why it matters — You must develop a "cost-benefit" filter for AI usage. If a task takes 30 seconds to do manually but 2 minutes to prompt and refine, the AI is a net loss. In the book — Stern is refreshingly honest about the "friction" of certain tasks, noting that for some simple, singular actions, the human brain is still the fastest processor available.

AI as a Sparring Partner

What it is — Using AI not to give you the answer, but to challenge your thinking. Why it matters — This moves AI from a "tool" to a "collaborator." It helps break through writer's block and identifies blind spots in your logic. In the book — She describes using AI to "argue" against her ideas, asking the AI to find flaws in her business plan or to provide three counter-arguments to a specific point of view.

The Human-in-the-Loop Necessity

What it is — AI lacks "lived experience" and "taste." The human must always be the final editor. Why it matters — AI can hallucinate facts and produce "uncanny" prose that feels hollow. You are the guardian of truth and the curator of style. In the book — Stern emphasizes that while the AI can generate a meal plan or a workout routine, the human must provide the "soul"—the preference for certain flavors, the physical limitations of the body, and the emotional motivation to keep going.

Frameworks & tools

The "Context-Task-Constraint" (CTC) Framework

To get high-quality outputs, Stern suggests structuring every complex prompt using these three pillars:

1. Context: Who is the AI? What is the situation? (e.g., "You are a senior marketing manager at a boutique skincare brand.")
2. Task: What exactly do you want it to do? (e.g., "Write a 3-month social media content calendar for a new hydrating serum.")
3. Constraints: What are the boundaries? (e.g., "Do not use emojis. Keep all captions under 50 words. Focus on the ingredient 'Hyaluronic Acid'.")

The Iterative Refinement Workflow

Instead of one-and-done prompting, use this 4-step loop:

1. The Seed: Ask for a broad outline or a "brain dump" of ideas.
2. The Selection: Pick the 20% of the output that is actually useful.
3. The Deep Dive: Ask the AI to expand on only those specific points.
4. The Polish: Ask the AI to rewrite the final result in a specific voice (e.g., "Write this in the style of a witty, slightly sarcastic New Yorker columnist").

Memorable examples & stories

The "Over-Planned" Vacation

Stern attempted to use AI to plan a complex, multi-city itinerary. The AI provided a massive, exhaustive list of every possible museum and restaurant. The Lesson: AI can suffer from "over-generation." It is often better to tell the AI to "give me 3 options" rather than "give me everything," to avoid information overload.

The Creative Sparring Partner

She used AI to help brainstorm a story idea she was stuck on. Instead of asking for the ending, she asked the AI to "tell me what a villain in this story would do." The Lesson: AI is incredible at "lateral thinking"—helping you see a different angle on a problem you've been staring at for too long.

The "Uncanny" Email

Stern tried to have AI write a sensitive email to a friend. The result was technically perfect but emotionally "off." The Lesson: For high-stakes emotional communication, AI should be used for structure (e.g., "Help me organize my thoughts for this email"), but the actual words must remain human.

Key quotes

"AI is not a replacement for your thinking; it is a laboratory for your ideas." (Paraphrased) Why it matters: It shifts the focus from 'AI replacing me' to 'AI as a tool for exploration.'

"The quality of the output is a direct reflection of the clarity of your intent." (Paraphrased) Why it matters: It places the responsibility for success on the human's ability to communicate.

"You have to learn to speak 'LLM'—which means being specific, providing context, and iterating until the result matches your internal vision." (Paraphrased) Why it matters: It highlights that prompting is a skill that requires practice and linguistic precision.

How to apply it

This week

1. The "Drafting" Experiment: The next time you have to write a long, boring email (like a project update or a request for a refund), write a "messy" bulleted list of what you want to say. Paste it into an AI and ask it to "Turn this into a professional, polite email."
2. The "Sparring" Experiment: Take a problem you are currently stuck on. Ask the AI: "I am trying to [Goal]. What are three reasons why this might fail, and how can I prevent them?"
3. The "Persona" Test: Next time you ask for a recipe or a workout, start the prompt with: "Act as a [Professional Title, e.g., Michelin-star chef] with 20 years of experience." Compare the result to your usual prompts.

Ongoing

• The 80/20 Rule of AI: Use AI for the first 80% of the work (research, outlining, drafting, brainstorming) and reserve the final 20% (fact-checking, tone-polishing, personal touches) for yourself.
• The "Context-First" Habit: Never ask a question without providing context. Instead of "How do I bake a cake?", try "I have a beginner's kitchen and want to bake a chocolate cake that is moist but not too sweet. Give me a simple recipe."
• The Iteration Mindset: Never accept the first response as the final answer. Always ask: "Can you make this shorter?" "Can you make this more humorous?" or "Can you explain the third point in more detail?"

Common pitfalls

• The "Magic Button" Fallacy: Many people get frustrated because they think AI should "just know" what they want. It doesn't. It only knows what you tell it.
• The Accuracy Trap: AI is a "prediction engine," not a "fact engine." It is designed to predict the next most likely word in a sentence, which means it can confidently state falsehoods. Always verify dates, names, and specific citations.
• The "Generic" Trap: If you give a generic prompt, you get a generic, "middle-of-the-road" response. To get "great" results, you must provide "great" constraints.

Connects to

• "Show Your Work!" by Austin Kenney: Connects to the idea of the human remaining the "editor" and the importance of the process over the final output.
• "Deep Work" by Cal Newport: Contrasts with the "shallow" tasks AI excels at, emphasizing that AI should be used to clear the "shallow" work so you can spend more time in "Deep Work" states.
• The "Centaur" Model of AI: The idea that the best results come from a hybrid of human and machine (Human + AI) rather than one replacing the other.

If you remember only three things

• AI is a collaborator, not a search engine: You must provide context, personas, and constraints to get high-quality results.
• The Iterative Loop is the secret sauce: Never settle for the first draft; treat the AI as a partner you can refine and push back on.
• Human-in-the-loop is non-negotiable: AI lacks taste, lived experience, and true accountability. You must always be the final editor.

Internalize it

• Reflection questions

  • Which part of my daily routine feels like "drudge work" that would be perfect for an AI to handle the first draft of?
  • When I get a bad result from an AI, was it because the prompt was too vague, or did I fail to provide enough context?
  • How can I use AI to challenge my own biases or "blind spots" in my current projects?
  • If I outsourced 50% of my administrative tasks to an AI "intern," what high-value creative work would I do with that extra time?

• Flashcards Q: What is the difference between a "Search" mindset and a "Collaborative" mindset in AI? A: A Search mindset expects a perfect answer from a query; a Collaborative mindset treats the AI as a partner that needs context, a persona, and iterative feedback.

Q: What are the three components of the CTC Framework? A: Context, Task, and Constraints.

Q: Why is the "Iterative Loop" important? A: Because the first response is often just a "first draft." Iteration allows you to refine the output, narrow the focus, and reach the specific nuance you need.

Q: What is the "Friction of Automation"? A: The realization that for some simple tasks, using AI actually takes more time than doing it manually. You must choose your battles.

Q: Why must a human always be "in the loop"? A: Because AI lacks lived experience, true taste, and the ability to verify facts perfectly. The human must be the curator of style and truth.

Q: How can AI be used as a "Sparring Partner"? A: By asking it to argue against your ideas, find flaws in your logic, or provide counter-arguments to your current perspective.

Q: What is the "Generic Trap" in prompting? A: Providing a generic prompt results in a generic, "middle-of-the-road" response. To get unique results, you must provide unique constraints.

Fidelity note

I have a high level of confidence in the core themes of this summary, specifically the "Collaborative" vs. "Search" mindset and the practical application of prompting (Context, Task, Constraints). The book is a personal narrative of experimentation, so while the methods (like the CTC framework) are standard best practices in the AI community, they are presented by Stern as the fruits of her year-long experimentation. I have ensured that the "Friction of Automation" and "Sparring Partner" concepts are central, as they are the hallmarks of her specific narrative. Readers should verify specific "AI tools" mentioned, as the technology evolves rapidly.