- DSTLLD
- Posts
- You Might Be Wasting Hours Of Your Life (Extended Distillation)
You Might Be Wasting Hours Of Your Life (Extended Distillation)
Cut down on busy work.
I’m going to do a little bit of evangelizing this week. I don’t feel like I’ve stepped up on my soapbox in a while and the last couple of weeks have shown me the true power of what I’m about to tell you.
In 3 days last week — in my spare time — I wrote almost 80,000 words.
I wrote the equivalent of an adult fiction novel in 3 days. Look out Stephen King.
We talked a few months back about escapism. In there I talked about how you want your escapes to be creative and social. Well, turns out I realized I hadn’t been doing much of either lately so I called up some old friends and we’re starting a new Dungeons & Dragons campaign in a few weeks.
I’m the dungeon master, which means a lot of time prepping. And, in my case, building the world my friends are going to play in, hence 80,000 words.
If you didn’t know, I’ve got a lot going on in my life right now with a new family, work and travel.
Not a ton of time to spend daydreaming about dragons, druids and dire wolves.
Enter AI.

I’ve been using AI for years — my particular choice is ChatGPT because of the interface and other integrations but just pick one and stick with it, the differences today are minimal — but I’d always had some cognitive dissonance about using it. It felt cheap. Like I wasn’t working hard enough. Or that the thoughts and writing it was producing weren’t my own.
I’ve now arrived at a different conclusion and here’s why:
After examining hundreds of AI outputs over the years I can see the direct causal link between the forethought I put into my prompt and the result produced by the AI.
What does that mean?
It means I’ve bucketed — a well prompted — AI into 2 categories. It is either a machine that speeds up your thought experiments or it’s an intern that does work for you 1000 times faster than a human could.
Note, a well prompted AI. The less information and context you give the AI, the less and less the resulting output will be related to your own thoughts.
Consider how I was getting AI to write me ad copy for this newsletter.
Prompt 1: Write me 5 Facebook ads for the DSTLLD newsletter using a PAS copywriting framework.
Prompt 2: I'm attempting to write ad copy for some twitter and Facebook ads. I have the following idea of what my message in the ad should be: Move from a dot collector to a dot connecter - this is the ethos of DSTLLD taking all the excess information and boiling it into useful wisdom - use this idea for possible ad copy alongside scavenger and forager analogy how we used to have to scavenge for information but now we have too much information and we need to forage for the correct or most useful stuff DSTLLD is finding signal in noise - it is refining 2 tons of ore into a drop of gold This needs to include information that the info they read will improve their life via self improvement. I cut through the noise in self improvement to give you the stuff that matters. I need to create ad copy with a PAS framework to deliver the most newsletter subscriptions. Give me 5 ad copy variations I could try.
Which prompt do you think produces better results and is more closely aligned with the vision I had in my head?
These are obviously specialized use cases. What happens if you don’t need to build a Dungeons & Dragons world or write ad copy. Is AI still useful for you?
YES!
Here’s a few more use cases to drive this home for anybody.
AI can teach you anything.
AI tutors are going to replace real tutors for your kids very soon, if they haven’t already.
Here’s a prompt to try:
Here is a framework for you: "It is important to view knowledge as sort of a semantic tree -- make sure you understand the fundamental principles, ie the trunk and big branches, before you get into the leaves/details or there is nothing for them to hang on to."
Teach me about [WHAT YOU WANT TO LEARN], but start with the roots and then work all the way up and don't move to the next layer until I say "I understand".
This semantic tree framework comes from Elon Musk and how he thinks about learning. I tried this with Systems Thinking a few weeks back and was able to grasp the fundamentals quickly.
Tired of reading online recipes?
I gave ChatGPT our meal preferences, dietary restrictions, examples of other things we liked and protein thresholds we were trying to reach. My wife is a vegetarian so protein is hard sometimes.
It gave me a 14 day meal plan. Generated novel recipes based on our preferences that we hadn’t tried before and created a recurring shopping list of everything we would need based on the recipes, quantities included.
Bonus: I get ChatGPT to examine online recipes and turn them into an Apple Note so I don’t need to read through Jennifer’s sob story at the start of every online recipe about how she found her recipe for risotto in her dead grandma’s diary 5 years ago.
I’ve heard of people doing the exact same thing with workout plans.
Audio Journal Entries
I call these my walk and talk sessions. In fact, the newsletter 2 weeks ago was written during one of these sessions on a walk, I just spoke the newsletter out loud. I fed the transcript of my audio note and a prompt to ChatGPT alongside one of my old newsletters that I liked and ChatGPT wrote 80% of that newsletter.
I use this for other purposes too. I find going for a walk and talking through my problems is usually a great way to solve them. Here’s the prompt I used the other day:
Attached is an audio transcription that was recorded while I was on a walk earlier. Please create an extended summary of all the key information I talked about while on the walk. Then include an executive summary of the 5 most important topics from the discussion, using your best judgement on how to improve my life. Finally create a list of actionable items based on the executive summary that I can add to my to do list.
AI isn’t going to solve all of your problems or produce perfect answers. But it might help you buy back a few hours of your day from now on. Don’t forget about the 10/80/10 principle.
Put in the effort to write a great prompt doing about 10% of the workload up front.
Have the AI complete 80% of the bulk of the work.
Tweak the final results with 10% more work yourself to really fine tune the vision you had in your head.
I hope I saved you some time.
Here’s ChatGPT’s summary of my newsletter if you wanted to save time 😉:
The author reflects on their recent productivity and highlights how AI can be a transformative tool for creativity, learning, and efficiency. Key points include:
AI-Enhanced Creativity:
In just three days, they wrote 80,000 words for a Dungeons & Dragons campaign, leveraging AI for world-building.
Emphasized that the quality of AI outputs depends on well-crafted prompts.
Applications of AI:
Creative Writing and Marketing: AI helps refine ideas and produce outputs like ad copy or creative content with speed and alignment to user intent.
Learning and Tutoring: AI can simplify complex topics using frameworks like Elon Musk’s "semantic tree" analogy, breaking knowledge into fundamental principles before diving into details.
Meal and Workout Plans: AI generates personalized plans, recipes, and shopping lists, saving time and effort.
Audio Journals: By transcribing spoken thoughts, AI helps organize ideas into actionable plans or polished content.
Efficiency Tips:
Advocated the 10/80/10 principle:
10% effort upfront to craft a detailed prompt.
AI handles 80% of the work.
10% final tweaks for refinement.
Mindset Shift:
The author overcame initial doubts about AI feeling "cheap" and now views it as a powerful accelerator for thought experiments and creative endeavors.
In essence, the newsletter is a call to embrace AI as a tool to reclaim time and enhance productivity across various aspects of life.
Words I Wish I Wrote
“If we have learned one thing from the history of invention and discovery, it is that, in the long run—and often in the short one—the most daring prophecies seem laughably conservative.”
Did you learn something new? If you enjoyed this, you can support DSTLLD by taking a moment to:
❤️ Like this post to get it in front of more curious people.
💬 Reply or comment to share your thoughts or ask a follow up question.
📤 Forward this post to someone who’d benefit from it.
Reply