
TL;DR: Today we’re releasing a new episode of our podcast AI & I. I go in depth with Evan Armstrong, the lead writer at Every. We get into how we use AI in the four stages of our writing process: having good taste, choosing a topic, the craft of writing, and building an audience. Watch on X or YouTube, or listen on Spotify or Apple Podcasts.
Evan Armstrong and I are publishing some of the best work of our lives—and we’re using AI to do it.
In this episode, we conduct an expert workshop on how language models feed our obsession with words and help us write for more than 78,000 readers every day.
Evan is the lead writer at Every who explores profit and power in technology in his twice-weekly column Napkin Math. We’ve been working together for three years, and I think Evan is smart, funny, and ineffably curious about the intersection of business and technology. He has the rare combination of business acumen, way with words, and crazy that is required to be a professional writer.
Evan is teaching a course on how to write with AI later this month, and we structured this conversation around the four segments of the class:
- Having taste—understanding what good is
- Picking a topic—knowing what to write about
- Crafting an essay—everything from sketching out an outline to editing
- Building an audience—learning how to reach people
As we talk, we use AI to identify my taste in books and movies live on the show. We also discuss how Claude is integrated into my research process, screen sharing through a Project I created for a long-form essay that I’m currently working on, and my workflow inside the LLM to come up with relevant metaphors. Evan takes me through the routine that makes it possible for him to consistently publish two essays every week, and screen shares the way he pushes through writer’s block with AI-powered word processor Lex and uses research tool Consensus to master new topics quickly.
I came out of this conversation learning more about how I can use AI in my process and I’m certain you will, too. Here’s a link to the transcript of this episode.
This is a must-watch for aspiring writers, or anyone whose job involves writing more than six sentences in a row.
Watch on X or YouTube, or listen on Spotify or Apple Podcasts.
If you want a quick summary, here’s a taste for paying subscribers:
The power of taste
In a piece a few months ago, Evan wrote, “[K]nowing what to make is just as important as the ability to make it.” The same applies to writing. “Everyone who wants to write something good has to be able to say what good is,” he says, and taste, or “being able to accurately describe why you liked a thing besides ‘awesome’ or ‘loved it,’” is a “discernible skill.” Here are a couple of Evan’s thoughts around taste:
TL;DR: Today we’re releasing a new episode of our podcast AI & I. I go in depth with Evan Armstrong, the lead writer at Every. We get into how we use AI in the four stages of our writing process: having good taste, choosing a topic, the craft of writing, and building an audience. Watch on X or YouTube, or listen on Spotify or Apple Podcasts.
Evan Armstrong and I are publishing some of the best work of our lives—and we’re using AI to do it.
In this episode, we conduct an expert workshop on how language models feed our obsession with words and help us write for more than 78,000 readers every day.
Evan is the lead writer at Every who explores profit and power in technology in his twice-weekly column Napkin Math. We’ve been working together for three years, and I think Evan is smart, funny, and ineffably curious about the intersection of business and technology. He has the rare combination of business acumen, way with words, and crazy that is required to be a professional writer.
Evan is teaching a course on how to write with AI later this month, and we structured this conversation around the four segments of the class:
- Having taste—understanding what good is
- Picking a topic—knowing what to write about
- Crafting an essay—everything from sketching out an outline to editing
- Building an audience—learning how to reach people
As we talk, we use AI to identify my taste in books and movies live on the show. We also discuss how Claude is integrated into my research process, screen sharing through a Project I created for a long-form essay that I’m currently working on, and my workflow inside the LLM to come up with relevant metaphors. Evan takes me through the routine that makes it possible for him to consistently publish two essays every week, and screen shares the way he pushes through writer’s block with AI-powered word processor Lex and uses research tool Consensus to master new topics quickly.
I came out of this conversation learning more about how I can use AI in my process and I’m certain you will, too. Here’s a link to the transcript of this episode.
This is a must-watch for aspiring writers, or anyone whose job involves writing more than six sentences in a row.
Watch on X or YouTube, or listen on Spotify or Apple Podcasts.
If you want a quick summary, here’s a taste for paying subscribers:
The power of taste
In a piece a few months ago, Evan wrote, “[K]nowing what to make is just as important as the ability to make it.” The same applies to writing. “Everyone who wants to write something good has to be able to say what good is,” he says, and taste, or “being able to accurately describe why you liked a thing besides ‘awesome’ or ‘loved it,’” is a “discernible skill.” Here are a couple of Evan’s thoughts around taste:
- Write what you’d read. Evan has discovered that his reading preferences are indicative of his own taste. “I only write things that I would have a good time reading…anytime I deviate from that…the audience doesn't like it, I don't like it, no one enjoys it,” he explains.
- Make the boring brilliant. Drilling down into what he enjoys reading, Evan seeks out writing that explains dry topics in an interesting, accessible way, adding that it reflects his aspirations as a writer. “It’s really hard to be accurate and enjoyable…entertaining reads about boring topics is how I think about it,” he says.
Evan and I also explore my taste together on this episode. My taste in books ranges from Robert Sapolsky-esque writing that makes hard topics accessible, to poetic, lyrical prose exemplified by Annie Dillard, and practical writing that offers actionable insights. I’ve come to realize that I’m drawn to interdisciplinary looks at human experience— like the relationship between humans and technology, or technology and creativity—bundled with psychology, philosophy, and business.
If I look back on my life, I can see this affinity toward the interdisciplinary emerge as a pattern in the things I’ve loved, but I was never able to articulate it until I used ChatGPT and Claude—and being able to put your taste into words is an important part of doing good creative work. I wrote about using LLMs to discover my taste in a piece a year ago, and it was one of my big unlock moments for AI. I go through the process I wrote about in the piece live in this episode, but the basic gist of the exercise is to make a list of people you admire, throw that list into ChatGPT, and prompt it to synthesize your taste.
Evan used this exercise to articulate his own taste, broadening it to encompass art forms beyond literature. His key insights include:
- Taste transcends mediums. Evan believes that it’s as important for writers to watch movies and go to art galleries as it is for them to read good books. “[W]hen you say, I like lyrical prose—how does that apply in other mediums? [Are] there lyrical paintings? My big thesis when it comes to taste is that it's a blob of emotional permission to like what you want to like and it's not constricted to certain types of medium,” he explains.
- Use AI to bridge art forms. Evan turns to ChatGPT to explore the different expressions of his taste. “I find that it’s able to [figuratively] draw things that I haven't heard of, or I haven't been interested in and it makes me more well-rounded…[for example, if] I’m reading something I really like, I’ll say, ‘Give me paintings like this,’” he says.
I like Evan’s approach because it highlights the part of taste that’s about exploring new things in different mediums that are like things that you enjoy. In his words, “I'm just so passionate about content and consuming good things…every day you should get to consume something that blows your mind—there's enough out there that you'll never run out…it's overwhelming to find, but [LLMs] make it a lot easier.”
The importance of topic selection
Choosing a topic is the part of Evan’s writing process where AI admittedly has the least intervention (and gym equipment admittedly plays a bigger role). This is an outline of his method:
- Sweat, chill, write. Since he publishes Napkin Math twice a week, and writes a large chunk of our weekly round-up, Context Window, Evan had to find a way “to go into the zone and pick the right thing” fast. This is how he does it: “Every morning after I do my lift, I go and sit in the sauna for 15 or 20 minutes…I think about what ideas are good…take a cold shower, come home, and write down the essay.”
- Disconnect to ideate. Evan’s process is a sign for writers who are always plugged in to take a break. “I spend all day jacked into technology…and if I try to sit down and write [my ideas] all out, I just get overwhelmed, but if I…just let it bubble up naturally, where I'm not thinking about anything, usually the first idea that comes into my head is the best one.
- Supercharge your research with LLMs. As Evan prepares to write an essay, especially about an unfamiliar topic, he uses Consensus to extract key insights about the subject. “[Y]ou eventually learn that if you read three papers, you've gotten to about 95 percent of the depth of most experts…as long as you're reading the right paper—and the issue…before LLMs was that it was really hard to know what the right paper was.”
The way I pick topics, on the other hand, is more heavily influenced by AI. This is more about my process, with quotes from this episode:
- Capture ideas on the go. I do a lot of my thinking while walking, so I either record a voice memo and feed the transcription into an LLM or talk to ChatGPT’s new Advanced Voice Mode, prompting the model to reflect my thoughts back to me. I find that it’s good at finding “an interesting thing” to dive deeper into from the “morass of things swirling around in my head.”
- Distill a clear thesis with LLMs. I’m working on a long-form piece that I’ve made pages of notes about, so I upload my research to Claude and prompt the model to write a cohesive thesis. “It's reflecting back to me patterns that it sees in what I've been thinking about and distilling it down…it's like a kaleidoscope where I get to look at all my notes from a different perspective.
The hard work of craft
This part of writing is about putting pen to paper. Evan writes inside Lex, which was incubated by Every and is run by Every cofounder Nathan Baschez. These are the ways he uses AI in this stage of his process:
- Cover your writing blindspots with AI. Evan has developed a custom prompt in Lex, trained on his previous essays, that generates 10 ideas of how he could conclude his essay based on his current draft. Whenever he finds himself struggling with a conclusion, he asks Lex “What would Evan say?” and the AI responds with a helpful jumping-off point.
- Polish your writing in Lex. The other Lex feature Evan finds helpful is “checks,” which reviews your writing and gives you suggestions to improve it. “The goal isn't to get rid of your editor, but the better and cleaner the draft is that you turn over to your trusted thought partner, the better the feedback they can give you,” he explains.
Here’s how I use language models while writing an essay, with quotes from this episode:
- Stay in the zone for longer with AI. I use AI in my writing to maintain momentum, using Lex to summarize familiar concepts without breaking my flow. When explaining a philosophy theory to my readers, instead of sifting through the internet, I rely on Lex to get “three or four sentences…that just explain exactly the idea,” which I’ll probably edit later, but is valuable because it’s “enough for me to keep going.”
- Delegate details to AI and keep moving. Another Lex feature I often use while writing is prompting the model to expand on my notes in specific sections of the article. “That’s just a really easy way to get the specifics in my head or get the examples I need so that I can keep going.”
- LLMs as a creative collaborator. While writing, I use Claude to create powerful metaphors, first by feeding it relevant research, and then prompting it for analogies. The process is a “dual collaboration,” where the model is “weaving in its own sense of what good is and what's relevant to create this metaphor that I probably wouldn't have come up with on my own.”
Finding an audience
The last component of the writing process is about getting the thoughts you put out into the world to reach people. At this stage, Evan often uses Spiral, the prompt builder that Every launched that automates 80 percent of repetitive writing tasks. Evan compares Spiral to the warp pipe in the Super Mario Bros. video game, where “sometimes he'll go into the pipe and then he comes out in like a different shape or he comes out in a different place…Spiral is a pipe designer—you can take one body of text and stick it in the pipe, and it transforms it into something else.” Here are his thoughts on building an audience:
- Put repetitive tasks on autopilot. Evan uses a Spiral fine-tuned to convert his essays into tweets to promote them on social media, efficiently addressing an important task that he finds unpleasant. “Social media is the bane of my existence…however, we do not make any money if people don’t find out essays, and so social media is a key component,” he explains.
- Engage the algorithm to reach the audience. Evan highlights that AI’s role in audience engagement goes beyond content creation, extending to being a part of the dreaded social media algorithm. “I think a really important point that most people miss is that you're serving the algorithm—the AI algorithm—as much as you're serving the audience in the context of distribution.”
Beyond tweets and Linkedin posts about AI & I or the articles I write, I find Spiral invaluable when I want to quickly tweet about new developments in AI. For example, I recently wrote a tweet about OpenAI’s new model Strawberry, with Spiral—and the process was easier and more efficient than if I had to write it myself.
You can check out the episode on X, Spotify, Apple Podcasts, or YouTube. Links and timestamps are below:
- Watch on X
- Watch on YouTube
- Listen on Spotify (make sure to follow to help us rank!)
- Listen on Apple Podcasts
Timestamps:
- Introduction: 00:01:04
- How to develop good taste: 00:04:28
- I use Claude to articulate my taste in books: 00:13:34
- How to use LLMs to explore art cross different mediums: 00:21:06
- The way Evan chooses his next essay topic: 00:33:45
- Go from research notes to clear thesis in Claude Projects: 00:38:20
- How Evan uses AI to master new topics quickly: 00:46:51
- Evan leverages AI to power through writer’s block: 00:59:21
- How to use Claude to find good metaphors: 01:04:28
- The role of AI in building an audience: 01:11:44
What do you use AI for? Have you found any interesting or surprising use cases? We want to hear from you—and we might even interview you. Reply here to talk to me!
Miss an episode? Catch up on my recent conversations with star podcaster Dwarkesh Patel, LinkedIn cofounder Reid Hoffman, a16z Podcast host Steph Smith, economist Tyler Cowen, writer and entrepreneur David Perell, founder and newsletter operator Ben Tossell, and others, and learn how they use AI to think, create, and relate.
If you’re enjoying my work, here are a few things I recommend:
- Subscribe to Every
- Follow me on X
- Subscribe to Every’s YouTube channel
Thanks to Rhea Purohit for editorial support.
Dan Shipper is the cofounder and CEO of Every, where he writes the Chain of Thought column and hosts the podcast AI & I. You can follow him on X at @danshipper and on LinkedIn, and Every on X at @every and on LinkedIn.
Ideas and Apps to
Thrive in the AI Age
The essential toolkit for those shaping the future
"This might be the best value you
can get from an AI subscription."
- Jay S.
Join 100,000+ leaders, builders, and innovators

Email address
Already have an account? Sign in
What is included in a subscription?
Daily insights from AI pioneers + early access to powerful AI tools
Ideas and Apps to
Thrive in the AI Age
The essential toolkit for those shaping the future
"This might be the best value you
can get from an AI subscription."
- Jay S.
Join 100,000+ leaders, builders, and innovators

Email address
Already have an account? Sign in
What is included in a subscription?
Daily insights from AI pioneers + early access to powerful AI tools