This is a blog post about AI. I wrote it on March 21, 2025, so hopefully it’s still relevant by the time you read it and there’s not like GPT 7o out ruling the world yet. 😬
Jokes aside, AI is anxiety-inducing for most knowledge workers.
Certainly it is for me.
When I was looking for stories of AInxiety the other day to see if I was not alone, I saw a redditor who put it well when they said something like:
“I get AInxiety due to this sense that there’s huge opportunity present, but that I don’t understand how to utilize it or capitalize on it, and meanwhile I feel the ground shifting beneath my feet and see the world of knowledge work changing.”
Freelancers feel this, course creators feel this — we all feel it.
I have some silver linings and strategies to share with you today that I believe will help you thrive in the world we’re finding ourselves in, but I just want to say right out of the gate:
If you’re feeling anxious, stressed, or overwhelmed, that’s 100% valid, and I feel it too.
I tend to chronically look for the silver linings and opportunities, and it can make people feel like they’re not seen if I’m only focused on that and not just able to say “that sucks.”
So I just want to say…
Yes, it’s stressful.
And threatening.
And that does, indeed, suck.
BUT.
There are also some huge opportunities available to us, and I have some ideas on how we can capitalize on them that I’d like to share with you today.
So let’s do eeeeeeeet
10 Predictions & Strategies For Freelancers to Survive Thrive in an AI World
These are just my beliefs — they could be wrong, and the nature of AI is that anyone making predictions about things is bound to look like an idiot in six months.
I’ve intentionally not published anything about AI so far in DYF, but the fact is, I care about you guys and I’d rather say something that becomes outdated than not share it out of fear and have you not being aware of some potentially-helpful gem for your biz.
So— here are some of my predictions and beliefs on how to thrive in this AI world as a knowledge worker.
…And they could very well be wrong or outdated and irrelevant by the time I click “publish.” 🙂
Belief #1 — AI-Augmented Output Is The New Table Stakes
For a minute there, the freelancers using AI to augment their fulfillment were the exception, and it gave them a competitive edge if they did a good job of it.
But for cutting-edge industries in particular (e.g. tech), I feel the winds-a-changing, and AI augmentation now feels like the new “table stakes” for commodity freelance work in particular.
Belief that I could be wrong about: The efficiency gains AI brings are soon going to be the expectation for commodity work — and the thing that sets the “commoditized market rate” — as more freelancers get access to AI tooling.
(This means it’ll make it even harder for commodity freelancers to scrape by financially — as if it wasn’t hard enough already.)
It feels to me like the era where commodity freelancers who say “I don’t use AI at all in my process” and can still do really well financially is coming to an end. (Maybe it has already, IDK)
Belief #2 — We Need To Shift ASAP From “Commoditized Implementation” Work → Into “Strategic Wrangling & Execution” Work
Belief that I could be wrong about: The more your work is strategic vs. purely technical, the harder it is going to be for an AI to replace you.
I believe that the “deciding” is what’s valuable in this AI landscape — that while the implementation is slowly getting consumed by AI, the “deciding” will be the last to go.
We can see this with the “vibe coding” wave of non-devs using AI tools to spin up apps incredibly quickly, but inadvertently pushing out security vulnerabilities, exploits, etc. because of their “unknown unknowns.”
Cool that they made an app quickly, sure. Not so cool that the way they made it is super flawed due to their newbness and that it’s going to blow up in their client’s face and maybe get their client sued when hackers can easily access all of their customers’ sensitive data because the vibe coder didn’t set up some super-basic security protocol that a more experienced dev would have known to roll out.
The era of the “commoditized specialist” might be coming to an end, but for freelancers who can align their expertise to strategically wrangle teams, AIs, automations, and tools to deliver valuable outcomes, they’re insulated by a natural “AI replacement moat.”
Belief #3 — Audit Your Secret Sauce
Take a hard look at what truly makes YOU uniquely valuable to clients. It’s probably not your “end result work” itself, but rather, how you personally get to the end result. The process you take. The decisions you make after interviewing your clients about their goals for the project. etc.
Ask yourself: “What parts of my process would be hardest for an AI to replicate?”
This is usually your strategic thinking, domain expertise, and client relationship management — not your implementation skills.
Protect these elements at all costs while being willing to let AI handle the aspects that don’t constitute your competitive edge.
If delegating the “grunt work” in your process to AI while retaining the strategic work allows you to deliver your outcomes in a fraction of the time, that’s a win in my book. (As long as you don’t charge by the hour anyway, ha ha. More on this in a minute.)
Your specialized workflows and systems are becoming some of your most valuable assets as AI commoditizes basic execution.
Belief #4 — Don’t Bury Your Head In The Sand Or Be Prematurely Dismissive of Tools / Models
When it comes to AI, I’ve noticed that Augmentation > Generation.
Freelancers who ask ChatGPT to “write me a good blog post” and then dismiss it – or dismiss AI tooling in general – when its output sucks as a result of their shitty prompt are setting themselves up to be outcompeted by other freelancers who take the time to really understand and explore the possibilities and limitations of AI tooling.
The winners aren’t blindly outsourcing their thinking — they’re finding the optimal human-AI collaboration that preserves their unique value while increasing efficiency.
Almost any complex task has components that can be AI-accelerated and components that should remain human-driven.
^ Those last 2 sentences there were AI-written, for example, and the 2 before were written by me. I had Claude write ’em with my own complex AI-writing-context-chain because I wasn’t sure how I wanted to close out the section. I won’t edit them, so that you can see subtle word choice differences, but normally I’d prob change a couple words.
Belief #5 — Delegate Properly To AIs, Build “AI Workflows,” And Keep Them Secret!
Belief that I could be wrong about: In the coming years, it will be increasingly the case that our competitive advantage as knowledge workers is our combination of expertise + our secret, behind-the-scenes, AI-augmented delivery workflow.
To help you find yours, apply the 10-80-10 delegation rule I learned from Dan Martell, ‘cept with AI instead of humans: Handle the first 10% yourself (strategic direction), delegate whatever you can of the middle 80% to AI (implementation), and own the final 10% (quality control and your distinctive finishing touch).
Create a novel combined “AI-chain” process that not just any ole sod with an internet connection could replicate. The easier your process is for a newb to stumble upon and replicate without quality loss, the less of a competitive moat you have.
Position your unique approach as part of your intellectual property, not as a commodity clients can buy once and replicate forever or that competitors can easily rip off to mimic your quality and output.
And be sure you harvest the gains of your process once you make it. If you charge by the hour and get 3x faster, you’re only screwing yourself because now you suddenly make a third what you did before from a project.
Don’t package your AI workflows as deliverables.
Don’t teach clients how to replicate your process.
Don’t just vomit some bullshit into chatgpt and blindly copy/paste its output or do whatever it says.
Be strategic. Be opinionated. Be discerning. Use your human brain.
If you do those things, you can achieve some pretty dang cool things with less effort than you’d otherwise spend. Sometimes not in terms of time, but rather in terms of energy and the cognitive drain of all the “deciding work.”
The goal here is deliver the same value you do now, for the same price you do now, but that you secretly behind the scenes were able to spend less time in order to do so.
(But this goes back to the importance of getting out of commodity work and solving valuable problems, because if you’re a commoditized freelancer on retainer, you probably don’t have the option not to charge by the hour.)
Belief #6 — Become Data-Driven & Discerning — And Position Yourself Around Your Outcomes
No longer does it take 10 years of practice & experience to spin up an app in an esoteric code language or make gorgeous designs.
Anyone can make an “80% good” thing in a stupid-short amount of time.
BUT the trouble is that they don’t know which 80% is good and which 20% sucks — and neither do the clients.
It’s important to define “what good is” and position yourself around your knowledge of that, and differentiate yourself from the others who can build an app in 2 days but not know what github is and ship some buggy trash without the client realizing it.
Your knowledge and experience is still your competitive advantage, but the way you position it is different.
Belief #7 — Look For The Potential To Do Things You Previously Couldn’t
Here’s what struck me the other day:
A LOT of the “businessy crap” and “management crap” that most freelancers don’t want to have to deal with, but previously kinda had to if they wanted to grow a proper business, is now becoming highly leveraged and automateable.
To the point that if you do things right, you can actually be a solopreneur AND run a very sophisticated business.
(Which you straight-up couldn’t do in the past)
This is AMAZING for freelancers who don’t want to run agencies or manage people.
You can do things now that in the past required entire departments of humans, in just a few hours a week.
Scary if you’re relying on being one of those humans in the department in order to make a living, but empowering if you want to be a freelancer doing interesting work, while having maximum freedom and not having to manage people.
Belief #8 — You Need To Control Your Lead Gen FIRST, Before You Can Truly Leverage AI
In my experience, huge AI-driven efficiency gains rely on having consistent, repeatable processes to optimize.
If you’re reliant on random referrals sending you whatever odd jobs they think of, you know all too well that you end up with a chaotic mix of projects that are all one-offs and look different from each other.
No repetition means no systematization, and no systematization means minimal AI leverage available.
This was true for SOPs and delegation in the pre-AI era, and it’s true for complex “AI automation chains” now.
Why I say that:
I’ve noticed that it usually takes me 3-6x the time it takes to “do a thing” to build an effective AI workflow that’ll get me a 70%+ efficiency boost.
What I’ve learned is that when you’re really trying to nail down quality and you’re not content for whatever bullshit ChatGPT spits out, it takes a lot of time to dial in a prompt and context-feeding process.
This 3-6x time cost means my “payoff period” comes after I’ve had to “do that thing” at least 4-8 times in the future, or whatever the math is there.
No-brainer for things I do 5 times a week. Not so nifty for things I do twice a year.
Point is: it’s often only worth it to systematize and automate processes you repeatedly do — and this is what most freelancers are missing.
If every client project is completely different (which happens when you rely on random referrals), you can’t build effective AI workflows or automations around your business.
The freelancers who will absolutely crush it in the AI era are the ones who own their lead generation and choose exactly who they work with.
When you control who comes into your pipeline, you can build repeatable systems around specific client types and problems — creating the perfect conditions for AI to supercharge your delivery.
This is why getting your lead gen figured out isn’t just about getting clients — it’s also key for creating the foundation that makes everything else in your business AI-augmentable.
And if you’re as concerned about becoming “AI-proof” as I am, this feels like a pretty non-negotiable thing to get sorted out.
Belief #9 — Non-Spammy Cold Outreach Is The Perfect Lead Gen Strategy For The AI Era
I’ve become absolutely convinced that value-forward, non-spammy, strategic cold outreach is the ideal lead generation approach for freelancers in the AI age.
Why? Three main reasons:
First, you get to hyper-target your ideal clients and ignore everyone else. This precision means you can build workflows and systems specifically for their problems.
Second, it lets you position yourself as a strategic problem-solver rather than just an implementation person — which is exactly what you need to avoid AI commoditization.
Third, cold outreach itself benefits enormously from AI augmentation. You can use AI to build targeted lists, scale personalization, and analyze response patterns while maintaining the human touch where it matters.
The combination is powerful: cold outreach helps you attract exactly the right clients for your strategic expertise, and AI helps you execute that outreach more efficiently than ever before.
This is precisely why I built the $200KF approach around cold outreach systems — not because I’m some cold email fanboy, but because it creates the perfect conditions for freelancers to thrive in this new AI world.
(Btw, Claude AI wrote that line above too, which made me chuckle, lol. I’ve got him trained on my brand voice well, I guess.)
If you’re going to mess with cold outreach, there are a lot of critical things you have to get right otherwise it won’t work. (And most freelancers mess up a few of them and fail as a result)
But if you do all those little things right, the results can be incredible. ($200KF covers all the little things btw, in case you were thinking of joining)
Belief #10 — The Fundamentals Haven’t Changed; The Way We Do Them Has
I remind myself of the “Leads, Offer, Sales” mantra all the time. If you don’t know it, it’s this…
I believe that if you want a “real business,” you need a strategy for 3 things:
- Leads — How you get in front of people who you can help.
- Offer — How you actually help them. The thing you do, and the outcome it delivers.
- Sales — How you show them you’re worth hiring.
None of the above has changed with AI.
None of the above will change with AI. (Unless things in the world go really really well and we live in a universal basic income utopia or things go really really not-well and we’re clubbing each other over the heads for water and toilet paper…)
For as long capitalism exists as a concept, so too will the 3 facets of a business.
They just look different now. The tooling and methods have evolved, but the fundamentals of what makes a successful freelance business remain constant.
You always need a way to get in front of people. You always need to have a valuable thing you can do for them. And you always need to have a way to show them it’s valuable.
…Unless they’re chasing you trying to club you for your toilet paper and water in a post-apocalyptic wasteland. In those cases you should probably just run and throw toilet paper rolls at them until they stop chasing you.
Point is, AI doesn’t change the need for these business pillars — it just changes how we approach building and executing on each of them.
Belief #11 — $200KF Is Useful For Setting You Up To Thrive In The AI Freelancing World
I’ve been heads-down for over a year now on a course called $200k Freelancer, and when I started work on it I didn’t set out to make an “AI For Freelancers” course — I set out to make a lead gen course.
The goal was “show freelancers how to get a predictable, controllable lead gen system.”
And I succeeded at that. The first beta cohorts are getting early results now and my confidence in the repeatability of the process gets stronger with each win.
I still don’t consider it “done,” yet despite now being about 1,400 hrs deep on it + making a complimentary SaaS for it to help students succeed faster + another 1,000 hrs in it from team DYF, but alas, perfectionism at work.
(The 6-month 200% ROI guarantee I rolled into $200KF was as much for me as it was for students, because it helped me get past my “what if the process only worked for me” imposter syndrome knowing that they were protected if it turned out the process only worked for me or something)
We’re in the final beta now, but it’s looking like after this last beta cohort, I’m on track for bulletproof rock-solid confidence in the thing. And maybe actually possibly considering it finished (or at least “finished-ish” FFS) There are only so many case studies my imposter syndrome can take while still making me doubt myself after all, lol.
But anyway, in making this lead gen course, I also accidentally sort of kinda created an “AI for freelancers course” as a result of it being such an AI-aligned course. The AI prompts, tools, and workflows I built into it to help students trickled down into the way it teaches you about harnessing AI in other ways for your business too. Here are some examples:
1: Automations — I ended up building in a whole extra mini-course into $200KF on how to use Make automations for lead generation purposes — and understanding how automations work is a HUGE boost in your ability to scale your business while maintaining quality.
2: AI Writing — I’ve learned that if I create a “themeplate” for an AI that explains strategy + feed it an example + have a solid prompt, the output can be incredible. You’ll get to leverage this directly through the cold email themeplates and rewriting prompts, but also learn the fundamental “themeplate rewriting approach” that’s applicable to tons of different aspects of your marketing and fulfillment (even for non-writers, btw)
3: Lead Generation — AI is baked in to the $200KF process, and you’ll learn how you can use AI to find your perfect ideal clients at scale.
4: Sales — ChatGPT won’t go on sales calls for you (yet) but it IS useful for helping you plan your sales process, call templates, funnel flow, offer stack, etc. in addition to my instruction & direct support.
5: Differentiation & Productization — The hardest part for many students in $200KF is the very beginning: Choosing a niche, problem, solution, and offer. These are the bedrock of effective marketing, and each step presents an opportunity to get stuck. But there are AI-assisted brainstorming prompts for each of those stages — a student named Vlad told me it felt like the prompts were like “being handed the keys to a Ferrari” for how it helped him pick a niche after struggling in the past.
6: Good AI Prompting — I’ve leveled up a lot in my prompting skills & AI mastery in the process of making this course. I don’t consider myself an “AI guru” but I think I probably know more than most course creators or freelancers at this point. I initially did all this leveling up to create solid AI prompts just to directly help you, because I knew a lot of devs would struggle with copywriting, niching, etc., and needed to find a good way to leverage AI to help maintain their output quality. BUT what’s happened is students also get a “double benefit” because you seeing how I’ve created my prompts that you get to use also helps you understand how to create good prompts for your own purposes in your business.
7: Delegation Skills — In the past, only the freelancers who wanted to scale an agency needed to learn how to delegate. Not anymore. I’ve noticed that the same skills that help you delegate to humans effectively also help you delegate to AI effectively. Nifty for you that I’ve got a mini-course inside $200KF on delegation too 🙂
If building skills in the AI stuff above sounds cool, AND you want to get your lead gen sorted and start getting ahead of the AInxiety all in one go, you’d probably enjoy the $200k Freelancer program.
We regularly sell out, so I’d recommend getting on the waitlist if you think you’re keen on joining next time we open. (Waitlisters get first dibs)
“What’s $200k Freelancer?”
It’s my “course magnum opus.” Over 1,500 hours on my end to develop it and another thousand or so from team DYF. I can confidently say it’s almost definitely the most comprehensive freelancer lead gen course on the market.
What It Does:
$200KF Gives you a predictable system for getting unlimited clients.
(70% of students who send their first campaign get leads from it & calls booked from it — and that’s only the beginning!)
Three Cool Things About $200KF:
1 — Comes with a 6-month 200% ROI guarantee to remove the risk for you.
2 — Also comes with templates, automations, AI prompts, and free bundled access to an exclusive SaaS platform I made to guarantee your success and automate the hardest steps.
3 — The ROI guarantee means that if it doesn’t get you clients, your enrollment fee’s on me.
Be sure to hop on the waitlist if this sounds cool 🙂