When I sit with founders today and review their content, I can almost always tell within 30 seconds who is still playing the 2015 game and who understands the AI reality we’ve walked into.
The painful truth is this: if your content can be written by a capable intern with a good prompt, it will be drowned by AI within months—if it hasn’t been already.
After 20 years of working with founders, CMOs, and brand leaders, I’ve learned that the winners aren’t the ones who “do more content.” They’re the ones who deliberately create things AI cannot convincingly fake.
Let’s walk through what that actually means in practice.
-
Stop publishing “information.” Start publishing insight.
Most founders I meet are still trying to “explain” their market: what it is, why it matters, how it works. That used to be valuable. Today, a generic AI model can do that in seconds.
Information is now cheap. Insight is not.
Information is: “Here are 5 trends in e-commerce logistics.”
Insight is: “Here is the trend that will quietly kill 40% of mid-market e-commerce brands in the next five years—and how I’d bet against it.”
One of the best examples of this shift is Stripe. Their content isn’t a list of “payment tips.” Their essays break down how the internet economy actually works, using data only they see and patterns they’ve spotted across thousands of companies. It’s not information; it’s a worldview backed by proprietary observation.
Ask yourself, ruthlessly: would a smart reader learn anything from your content they couldn’t get by spending 5 minutes with ChatGPT?
If the answer is no, you’re not doing content marketing. You’re just adding noise.
-
Put “uncopyable data” at the core of your content
AI is trained on the public internet. That means anything common, public, or generic can eventually be mimicked.
What it cannot fabricate convincingly is your:
-
Proprietary data
-
Private customer patterns
-
Operational learnings
-
Failed experiments
Look at what Airbnb has done over the years. Their most compelling narratives weren’t generic travel guides. They leaned heavily on host data, guest behavior, and booking patterns. They could say things like, “In cities where hosts add this one sentence to their listing, bookings go up 18%.” That’s uncopyable, because only they see it at that scale.
You probably have far more uncopyable data than you realize:
-
Sales conversations your team has every week
-
Support tickets and what they reveal about real-world use cases
-
Internal performance metrics across customer segments
-
Post-mortems of deals you lost
Most founders let this gold rot in internal docs and Slack threads.
If you want to win in an AI-saturated world:
-
Turn your CRM into a content engine: “What we learned from losing 40 deals in Q2”
-
Turn your helpdesk into a pattern radar: “Why 60% of our churn came from one overlooked onboarding mistake”
-
Turn your feature usage metrics into market insight: “The feature everyone requested but almost no one used—and what that says about our industry”
An AI can sound smart, but it cannot independently discover what’s actually happening inside your product, with your customers, in your specific niche. That’s your edge—if you decide to use it.
-
Bring back the first-person founder narrative (for real, not as a gimmick)
Founders often tell me, “We want to sound professional, not personal.” What they usually end up sounding like is: forgettable.
The brands that cut through now have something AI can’t fully counterfeit: a consistent, recognizable human voice tied to real decisions, real risk, and real accountability.
Think of how Elon Musk has shaped the narrative around Tesla and SpaceX. You may love or hate his style, but you can’t say it’s generic. Every comment, every tweet, every off-the-cuff remark creates storylines AI wouldn’t dare manufacture on its own—because they’re rooted in real commitments and consequences.
Or look at how Brian Chesky talks about Airbnb’s hardest moments: regulatory fights, pandemic freefall, layoffs, product reinventions. It’s not “Our company faced challenges.” It’s “We sat in this room, looked at this data, and agonized over this decision.” Very specific. Very human.
If you’re a founder, your content should sound like:
“I believed X in 2018. I was wrong, and it cost us Y. Here’s what we changed, and here’s what I believe now.”
I’ve seen this pattern again and again:
-
Early-stage founders hide behind “we” and corporate tone, hoping to look bigger.
-
Late-stage founders desperately try to rehumanize their brands after years of sterile content.
Skip that mistake. Start with your actual voice. AI can clone writing style, but it can’t credibly clone lived decisions, personal regrets, and earned convictions.
-
Share your “earned secrets,” not surface-level takes
Over two decades, the most valuable content I’ve seen from founders has one thing in common: it reveals something they only learned by doing the hard, boring, or painful work themselves.
I call these “earned secrets.”
They’re not necessarily dramatic. Often they sound almost obvious—but only after someone shows you the pattern.
For example:
-
A DTC founder who realized 80% of their “brand” problems weren’t creative at all, but warehouse reliability and delivery consistency. Their most popular content? A deep dive called, “Brand is built in your logistics, not your Instagram.”
-
A SaaS CEO who realized the most accurate predictor of long-term retention was not company size or budget—but whether the buyer personally did a manual workaround before buying. That insight turned into a killer piece: “The hidden pre-purchase behavior that made our best customers.”
These are not things you get from scraping the internet or remixing blog posts. They came from pattern-recognition across hundreds of real cases.
Ask yourself:
-
What do I now know that I absolutely did not understand in my first 3 years?
-
What do we see in our data that contradicts “common wisdom” in our industry?
-
What would make a peer founder say, “Wait, I need to screenshot this”?
If your content doesn’t make someone in your exact audience stop, think, and possibly revise a decision or belief, it’s not insight yet. It’s commentary.
-
Move into formats that are hard for AI to fake
Text will be the easiest medium for AI to overrun. It already is.
So you need a portfolio of “un-AI-able” or at least “hard-to-AI-well” formats:
a) Live video, not just polished clips
Live Q&As, office hours, live product teardown sessions, unscripted debates.
When Figma’s founder and team do live walkthroughs or AMA-style sessions with designers, they create moments AI can’t plausibly recreate because:
-
Real people ask unscripted questions
-
There’s no perfect smoothing of language
-
You see hesitations, clarifications, thinking in real time
b) Audio communities, not just podcasts
Everyone wants to “start a podcast.” Few build real audio communities.
The real power is when:
-
Listeners become regular contributors
-
Customers show up live
-
The founder gets known for recurring, live, opinionated conversations
Think about how some VCs and operators turned Twitter Spaces or live audio rooms into weekly rituals. They weren’t just pushing episodes; they were hosting a room where people showed up for the unscripted friction of live debate.
c) Physical and curated experiences
I’ve seen more traction from a single well-run, small in-person workshop than from six months of “content calendar” output.
Patterns I’ve watched work:
-
A founder hosts a small, invite-only dinner for ten customers, then writes about the strongest insight (with permission and anonymization).
-
A CEO runs a hands-on workshop, then publishes the frameworks, the failures, and the unexpected questions that came up.
AI can summarize a conference. It cannot be the reason that ten specific people flew to a city to sit around a table with you.
The more you anchor your content in real moments, with real people, in specific contexts, the harder you are to imitate.
-
Build a founder brand that feels like a person, not a logo
Over the years, I’ve watched a consistent pattern: when a founder is visibly learning, arguing, admitting mistakes, and taking stands in public, the company doesn’t need to buy as much trust. It’s already there.
What does that look like tactically?
a) Human connection over polish
Share the messy bits:
-
How you navigated a critical product decision when your team was split
-
Why you killed a feature customers said they wanted
-
What you got wrong about hiring, culture, or go-to-market
Not in the “here’s a sanitized story with a perfect moral” way, but in the “here’s what still keeps me up at night” way. People are starved for that level of honesty in a world of AI-smooth narratives.
b) Opinionated stances (with receipts)
Most brand content is so neutral it falls straight through the brain. In an AI-saturated world, safe is invisible.
Look at Patagonia. Their stance on environmental issues is not an add-on; it is the brand. Their statements carry weight because they’re backed by decades of consistent, costly decisions.
Your version of that might be:
-
“We will not discount our product, ever, and here’s why.”
-
“We refuse to build X feature because we think it harms customers in the long run.”
-
“We believe this common industry metric is misleading, and here is what we track instead.”
AI can synthesize arguments; it cannot convincingly hold a line when that line costs money, reputation, or growth. You can.
c) Behind-the-scenes access
This is where I’ve seen the biggest gap between what founders could do and what they actually do.
Show:
-
Early Figma mockups, failed prototypes, or pitch decks that went nowhere
-
Your actual weekly leadership agenda and how it changed over time
-
The internal email you sent when something broke badly
As you do this, your audience starts to think: “I’m not just reading content; I’m watching a company being built in real time.” That’s deeply hard to fake and even harder to commoditize.
-
Optimize for “information gain,” not just impressions
One of the most expensive mistakes I see: founders chase views, not impact.
In the next wave of search and discovery, algorithms will get much better at identifying “information gain”—how much new, useful, non-obvious value someone got from a piece of content.
To set yourself up for that world, focus on three things:
a) New search engines (and new expectations)
As AI-powered search and answer engines rise, they will favor:
-
Specificity over generality
-
Original claims over rephrased consensus
-
Concrete examples and data over vague advice
If you write a piece titled “How to improve your onboarding,” it’s competing with thousands of almost identical pages and infinite AI rewrites.
If you write, “We cut our onboarding time by 47% by deleting these three steps,” that’s a different level:
-
Specific number
-
Concrete context
-
A claim that can be evaluated and quoted
b) Quote-worthiness
A simple test I use when reviewing content: does this piece contain at least one sentence that:
-
A peer founder would want to screenshot
-
A newsletter writer would want to quote
-
A conference speaker would want to reference
Think about how many people have quoted Amazon’s “It’s still Day 1” philosophy or “your margin is my opportunity.” Those are sharp, memorable mental hooks, anchored in how they actually operate.
You don’t need slogans; you need sharp, specific lines that package your thinking in a way others can carry.
c) Community distribution over algorithm roulette
The best-performing content I’ve seen in the last few years didn’t “go viral” in the old sense. It spread inside tightly defined communities:
-
Slack groups
-
Private founder circles
-
Niche industry forums
-
Customer communities
To win there, your content has to be:
-
So useful that someone risks their reputation to share it
-
So specific to a niche that it wouldn’t even make sense to a general audience
AI-generated content is rarely that specific, because it’s trying to please “everyone.” Your job is to be indispensable to the right 1,000 people, not mildly interesting to 100,000.
-
How to practically reset your content strategy in the next 90 days
If I were sitting with you in your office, here’s what I’d tell you to do, step by step:
-
Audit your last 12 months of content
-
Highlight anything that required proprietary data, real decisions, or personal risk to create.
-
Everything else is likely replaceable by AI. Be honest about that.
-
Identify 5–10 “earned secrets”
-
Bring your sales, support, and product leaders into a room.
-
Ask: “What do we know that our competitors probably don’t? What patterns do we see that outsiders wouldn’t believe at first?”
-
Turn each into a detailed outline backed by numbers, stories, and real examples.
-
Pick 2 “un-AI-able” formats and commit
-
Maybe it’s a monthly live founder Q&A.
-
Maybe it’s a small in-person salon every quarter and a write-up afterward.
-
Maybe it’s a recurring audio room with customers and peers.
Don’t try to do everything. Do a few things consistently and deeply.
-
Put your name on 3–5 pieces this quarter
Not as a “letter from the CEO” written by the marketing team, but content you’ve genuinely shaped:
-
Where you changed the headline because you cared
-
Where you added the uncomfortable story you weren’t sure you wanted out there
-
Where you’re willing to have your thinking challenged publicly
-
Design every piece for information gain
For each article, podcast, video, or thread, ask:
-
What will my exact target reader now understand, decide, or do differently because of this?
-
What sentence or idea here is most likely to be quoted or screenshotted?
If you can’t answer those questions, you’re not done yet.
-
The uncomfortable but liberating reality
The era where you could “just publish more” and win is over. AI will always be better at “more.”
But AI has a glaring weakness: it has no skin in your game.
It didn’t:
-
Spend five years failing to crack a market before finally finding the wedge.
-
Lay off people and feel sick about it for weeks.
-
Argue with co-founders at midnight over a product direction that could make or break your next 18 months.
-
Watch a seemingly small operational decision compound into a major brand asset—or a lingering liability.
You did. Or you will.
If you’re willing to turn those lived experiences, proprietary patterns, and real-time decisions into honest, sharp, specific content, you’re already playing a different game than 99% of the AI-generated noise.
In an AI-saturated world, your advantage is not that you can write. Your advantage is that you can know, decide, and stand for things an AI cannot.
Your content should prove that.