What I use AI for in my personal life, hopefully this will inspire and inform.

Cinematic cyberpunk street photography at night. A lone, individual sitting on a wet metal bench, intensely focused on a glowing, transparent digital tablet displaying dense, continuous text. The tablet casts a heavy pink neon glow on their face. In the background out of focus, a towering corporate skyscraper is covered in chaotic holographic bullet points and glitchy "TL;DR" text. Volumetric lighting cuts through the teal and magenta color grading. Shot on 35mm lens, f/1.8, sharp focus on the user and the tablet screen, grainy film texture, wet asphalt with high-contrast reflections. Hyper-realistic, 8k resolution.

Why I’m Resisting the “Summary” Habit

I have a confession to make: I think I’ve been letting AI make me a bit lazy.

I while back, I noticed that sitting down to read a long-form article or a 20-page report felt like a chore. My brain had become conditioned to look for the “TL;DR,” the bullet points, and the quick takeaways. I’ve been using AI to summarise almost everything, and while it felt like a productivity win in the moment, I’ve started to realise the hidden cost. We are losing the actual context.

The Rhythm of Short-Circuiting

When you ask an AI to “break this down for me,” you aren’t just saving time. You’re delegating your understanding. I recognised this happening a while back, and correcting it has been genuinely difficult. The irony is that the time AI saves you on tasks has to go somewhere, and for me, it’s had to go back into reading. Whether that’s checking what the AI produced or just building my own understanding of a topic.

I’m worried that if we get too deep into this rhythm, we’ll end up in a place where “no one really knows anything” anymore. We’ll all just be experts in reading bullet points generated by a model that might have missed the most nuanced part of the argument. I have said this a million times but I will say it again, I don’t want the models to do the thinking for me. I want them to help me move faster after I’ve done the heavy lifting.

My New “Deep Work” Workflow

To fight this, I’ve started making some intentional changes to how I work:

  1. The “Read-Through” Requirement: Even if I get a PDF summarised, I’ve made it a rule to sit down and read the full document first. I want to see the author’s intent and the data before the AI flattens it out.
  2. Back to Plain Text: I’m moving away from PowerPoint and back into plain text documents. There is something about a continuous flow of words in a Word doc or a text file that forces you to follow a logic chain from start to finish. It prevents you from “chunking” ideas into disconnected slides.
  3. Verify, Don’t Trust: I use AI to double-check my work, but I’ve stopped letting it start my work. Never take a summary at face value. I use the summary as a checklist to go back and verify specific points in the source text.

Don’t Offload The Thinking

Yes, this takes longer. I won’t pretend otherwise (but you have saved time in other places right?)

When everyone around you is hitting “Summarise” and moving on, slowing down to actually read something feels almost contrarian. But I’ve stopped framing it as a time cost and started seeing it as an investment in actually having something to say. The “aha!” moments, the unexpected connections, the opinion I form halfway through a chapter, none of that happens if I’ve only seen the digest version.

There’s a version of productivity that looks impressive on the surface but quietly hollows out your thinking. I don’t want that. I’d rather be slower and genuinely understand something than be fast and just vaguely aware of it.

Leave a comment