Why Founders Are Watching Nate Herk, Liam Ottley, and Mark Kashef

Why Founders Are Watching Nate Herk, Liam Ottley, and Mark Kashef (August 2026)

Most founders have tried at least three AI tools. They're still doing everything themselves.

The problem isn't the tools. It's that most AI content teaches you to use a chatbot to write faster emails. Meanwhile, a small group of YouTube creators is teaching something completely different: how to build AI systems that run entire parts of your business.

That's why Nate Herk went from zero subscribers to over 500,000 in less than two years. That's why Liam Ottley has 713,000 subscribers and a business model that hundreds of agencies have copied. That's why Mark Kashef has trained over 700 professionals on AI implementation and his videos get shared in founder Slack channels every week.

These aren't theory channels. They're showing you how to build the thing this weekend. And the founders watching them aren't looking for productivity hacks. They're looking for a way to scale their business without hiring first.

What Makes These AI Automation YouTube Channels Different

The fastest-growing AI YouTube channels in 2026 share one trait: they assume you're here to build something, not just learn about it.

Most AI content is surface-level. "10 ChatGPT prompts for coaches." "How to use AI to brainstorm ideas." It's helpful for about 15 minutes, then you're back to doing everything manually.

The channels founders are watching don't teach prompts. They teach systems. Multi-step workflows that connect tools, move data, make decisions, and run without you.

The format is step-by-step, screen-shared, and built for implementation. You can pause the video, copy what they're doing, and have a working automation by the end of the tutorial.

That's the difference. You're not watching someone talk about AI. You're watching someone build it, then you build it too.

Nate Herk: From Goldman Sachs to 500K Subscribers in Under Two Years

Nate Herk left Goldman Sachs to focus on AI automation full-time. That decision happened less than two years ago. By August 2026, he has over 500,000 subscribers.

That growth rate isn't an accident. Herk's content is built for people who want to automate a real business process this weekend, not theorize about AI's potential.

His specialty is n8n, an open-source workflow automation tool that connects apps, APIs, and AI models. He walks through building agents that do actual work: pulling data from one system, processing it with AI, and pushing results to another system without human intervention.

The format is methodical. He shows you the node setup. He explains why each connection matters. He troubleshoots errors on camera. You see the whole process, not just the highlight reel.

This is what founders need. Not inspiration. Instructions.

Herk's audience skews toward consultants, agencies, and expert service providers who are the bottleneck in their own business. They're revenue-generating, and they're tired of being the only person who can do the work.

His tutorials teach them how to build the systems that scale their business without adding headcount first. That's why his comment sections are full of people sharing what they built after watching.

Liam Ottley: The Creator Who Coined the AI Automation Agency Model

Liam Ottley has 713,000 subscribers as of August 2026. He's best known for coining the term "AI Automation Agency" and teaching a business model that hundreds of agencies have adopted.

The model is straightforward: agencies sell AI automation services to businesses that need workflows built but don't have the in-house expertise. The agency builds the agent, trains it, and hands over a system that runs on its own.

Ottley's content covers both sides of that equation. He teaches agencies how to find clients, scope projects, and deliver automation work. And he teaches the technical side: how to actually build the multi-step workflows that clients are paying for.

His approach is practical. He's not selling a dream. He's showing you the business model, the pricing structure, the client onboarding process, and the tools you need to deliver the work.

That transparency is rare in the AI space. Most creators sell courses without showing you what's inside. Ottley gives you enough in the YouTube video that you could start today.

His audience includes agency owners, consultants building automation practices, and founders who want to understand what's possible before they hire someone to build it. The comment section is full of people comparing notes on what worked and what didn't.

Ottley's content also covers tool selection, which matters more in 2026 than it did two years ago. The AI tool landscape changes constantly. Pricing shifts, features get deprecated, platforms shut down or pivot. Ottley keeps his recommendations current, which is one reason his audience trusts him.

Mark Kashef: Multi-Step Prompting and AI Agent Implementation

Mark Kashef has trained over 700 professionals on AI implementation. His YouTube content focuses on building AI agents and teaching multi-step prompting, the technique that turns a basic chatbot into a system that can handle complex tasks.

Multi-step prompting is exactly what it sounds like: breaking a task into smaller steps, then training the AI to move through those steps in sequence. It's the difference between asking AI to "write a proposal" and training it to first analyze the client brief, then pull relevant case studies, then draft an outline, then write the proposal, then format it for delivery.

Kashef's tutorials walk through that process in detail. He shows you how to structure the prompts, how to pass context from one step to the next, and how to troubleshoot when the AI misses something.

His content is particularly useful for teams and departments adopting AI together. HR and learning and development leaders watch his videos to understand what their teams need to learn. Department heads share his tutorials in internal channels.

Kashef's approach is less flashy than some creators, but it's grounded. He's teaching the fundamentals that make AI systems reliable. That's what professional teams need. They can't deploy a system that works 80% of the time. They need something they can trust.

His audience includes professionals who want to use AI to become indispensable at work, and leaders who are bringing AI to their teams and need a framework that works across skill levels.

Why Founders Watch These Channels Instead of the Big AI Names

The biggest AI YouTube channels aren't always the most useful. A channel with 3 million subscribers might be great for news and commentary. But if you're a founder trying to automate your client onboarding process, you don't need commentary. You need a tutorial.

That's the appeal of Herk, Ottley, and Kashef. They're not summarizing headlines. They're building the thing on camera and telling you how to replicate it.

The format is specific enough to implement, but general enough to adapt. You're not watching them build a system for their exact business. You're watching them build a system you can modify for yours.

This is the content gap that most AI creators miss. Founders don't need motivation. They're already motivated. They need instructions.

These channels also attract a high-value audience. Founders, agencies, consultants, and marketing teams. People who can afford tools, courses, and done-for-you services. That's why the sponsorship and course economics behind these channels work. The audience has buying power.

The Real Reason This Content Gets Millions of Views

The video counts tell the story. Herk's workflow tutorials regularly hit six figures in views. Ottley's business model breakdowns get shared across founder communities. Kashef's prompting guides get bookmarked and rewatched.

The reason isn't production quality. These aren't cinematic videos. They're screen recordings with clear audio and a structured walkthrough.

The reason is relevance. These creators are solving the exact problem their audience is stuck on right now.

Most founders hit the same wall: they've tried AI, they've subscribed to tools, and they're still doing everything manually because the AI doesn't know their business. It gives generic outputs that need heavy editing. It misses context. It can't make decisions. So they go back to doing it themselves.

The channels that are growing fastest in 2026 teach the missing step: how to train AI on your business, then build the workflows that use that context to do real work.

That's what Context Training is. Teaching your AI everything it needs to know to do the job you're asking, refining it as you go, so results get better over time instead of staying generic forever.

AI without your context is a brilliant stranger guessing at your business. These creators are teaching founders how to close that gap.

What These Channels Teach That Most AI Content Skips

The tutorials on these channels go deeper than prompts. They're teaching you how to build systems that connect tools, move data, and make decisions based on logic you define.

Here's what that looks like in practice. Say you're a consultant who sends a proposal after every discovery call. Right now, you're writing it by hand, pulling in past work samples, customizing the scope, and formatting the document. It takes two hours.

An AI automation for that process would: pull the call notes from your CRM, extract the client's goals and pain points, match those to your service offerings, draft the proposal using your structure and language, pull relevant case studies from your library, format the document, and drop it in your review folder.

You review it for five minutes, make small edits, and send. Total time: 15 minutes instead of two hours.

That's not a chatbot writing faster. That's a system doing the job.

The channels founders are watching in 2026 teach you how to build that system. They walk through the workflow logic, the tool connections, the prompts at each step, and the error handling when something breaks.

They also teach you how to test it. How to run the system on real data, catch mistakes early, and refine it until it's reliable.

Tools You'll See in These Tutorials

The tools these creators recommend show up across most of their tutorials. Not because they're sponsored, but because they're the tools that connect well and handle real business workflows.

n8n is the automation backbone. It's open-source, connects to hundreds of apps and APIs, and gives you full control over the workflow logic. Herk uses it constantly.

Make (formerly Integromat) is another automation platform that shows up often. It's visual, beginner-friendly, and handles complex workflows without code.

Claude and ChatGPT are the AI models doing the processing work. You'll see creators use both depending on the task. Claude tends to handle longer context better. ChatGPT is faster for quick tasks.

Airtable and Google Sheets show up as databases. They store the context your AI needs: client data, service offerings, past work samples, templates.

When voice matters, ElevenLabs is the go-to. It clones your voice and turns text into natural-sounding speech. You'll see it used in tutorials that automate video content, podcast production, or any workflow where you need audio that sounds human.

If you're publishing video content regularly, Opus Clip is the tool that cuts long videos into short-form clips for social. It's AI-powered, pulls the best moments, and formats them for each platform. You'll see creators mention it when they're teaching content repurposing workflows.

The Distinction That Separates Good Tutorials from Great Ones

Not every automation tutorial is worth your time. Some teach you how to complete a task. The best ones teach you how to build something that owns a role.

That distinction matters more than most creators realize.

An agent completes a task. An AI employee owns a role.

A booking agent that finds one speaking opportunity is doing a task. A Speaker Booking Agent that pitches you to five stages every week, tracks every reply, follows up with decision-makers, and owns your entire pipeline is an employee.

The creators getting the most traction in 2026 understand that difference. They're not teaching you to automate one email. They're teaching you to build a system that handles the entire job from start to finish.

That's what founders need. Because if you're still the bottleneck after implementing AI, you haven't solved the problem. You've just automated a step.

Why Strategy Comes Before Tools

The channels that grow fastest are the ones that teach strategy first, tools second.

Most AI content leads with the tool. "Here's how to use this new feature." But if you don't know what job you're trying to automate, learning a new feature doesn't help.

The better approach is clarity first. What job needs to get done? What does success look like? What steps are involved? Once you know that, the tool choice becomes obvious.

AI is the car. Clarity is the map. Without the map, you're driving fast in the wrong direction.

Herk, Ottley, and Kashef all teach this implicitly. They don't start with "here's a cool tool." They start with "here's the job we're automating" and then show you the tools that solve it.

That's the approach that works. And it's why their audiences implement what they learn instead of just watching and moving on.

What Founders Are Building After Watching

The comment sections on these channels are full of implementation stories. Founders sharing what they built, what worked, and where they got stuck.

Common use cases that show up again and again: client onboarding workflows, proposal generation, content repurposing systems, email follow-up sequences, data entry automation, reporting dashboards.

These aren't flashy projects. They're the behind-the-scenes work that eats time every week. The work that keeps founders busy but doesn't generate revenue.

Automating that work creates leverage. It frees up hours every week that can go toward the revenue-generating activities only you can do: sales calls, strategy sessions, high-value client work.

That's the outcome founders are chasing. Not "I saved five minutes on a task." But "I got 10 hours back this week and used it to close two more clients."

Why This Niche Has Strong Economics Behind It

The AI automation niche on YouTube isn't just growing in subscribers. It's growing in revenue potential.

These channels attract a high-value audience with real budgets. Founders who spend money on tools, courses, agency services, and consulting. That makes the niche attractive to sponsors and course creators.

The creators in this space aren't just making ad revenue. They're building education businesses, consulting practices, and productized services on the back of their YouTube audiences.

Ottley teaches the AI Automation Agency model and sells courses on how to start and scale that business. Kashef trains teams and offers implementation support. Herk's audience turns into consulting clients and course buyers.

The channel is the top of the funnel. The real business is what happens after someone subscribes.

That's why this content keeps getting better. The creators have a financial incentive to teach well, stay current, and keep their audience engaged.

What Makes a Tutorial Worth Watching in 2026

Not every AI tutorial is created equal. Here's what separates the ones worth your time from the noise.

First, it's step-by-step. You can pause the video and replicate what you're watching. If the creator skips steps or assumes you already know something, the tutorial falls apart.

Second, it's current. AI tools change fast. A tutorial from 2024 might be completely outdated by August 2026. The best creators update their content regularly or tell you when something has changed.

Third, it's specific. "How to use AI for marketing" is too vague. "How to build an AI workflow that turns podcast episodes into five LinkedIn posts, three Twitter threads, and a newsletter draft" is specific enough to implement.

Fourth, it assumes you're building for your business, not just following along for fun. The best creators explain why each step matters and how to adapt it to your use case.

Fifth, it's honest about what works and what doesn't. AI automation isn't magic. Sometimes the workflow breaks. Sometimes the output needs editing. The creators who admit that build more trust than the ones who oversell.

How to Use These Channels as a Founder

Here's how to get the most value from these AI automation YouTube channels without falling into the trap of watching tutorials forever and never implementing.

Start with the problem, not the video. Identify one task or workflow in your business that takes too much time or creates a bottleneck. Write it down. Be specific.

Then search for tutorials that solve that exact problem. Don't watch videos at random. You'll waste hours learning things you don't need yet.

Watch one tutorial all the way through before you start building. Take notes. Pause when you need to. Don't try to follow along in real time on your first pass.

Then build it. Open the tools, follow the steps, and create the workflow. Don't aim for perfect on the first try. Aim for functional.

Test it with real data. Run the workflow on an actual project or client task. Catch the errors early. Refine the logic until it works reliably.

Once it's working, document it. Write down what you built, why you built it, and how to maintain it. That documentation saves you hours later when you need to update or troubleshoot.

Then move to the next workflow. Don't try to automate everything at once. One working system beats five half-finished projects.

Why Context Training Is the Missing Step in Most Tutorials

The best tutorials teach you how to build the workflow. The best outcomes happen when you also train the AI on your business first.

Most automation tutorials assume your AI already knows enough to do the job. But it doesn't. AI models are trained on general knowledge. They don't know your clients, your voice, your service offerings, your past work, or your decision-making process.

That's why the outputs often feel generic. The AI is doing its best with incomplete context.

Context Training solves that. You teach your AI everything it needs to know before you ask it to do the work. Then you refine that context as you go, so the AI gets better over time instead of staying generic.

The channels that teach workflows are valuable. The next step is training the AI on your business so those workflows produce outputs you can actually use.

That's the lens that Seed & Society brings to AI education. The automation is the infrastructure. The context is what makes it yours.

How to Know If a Channel Is Actually Helping You Scale

The test is simple. After watching a tutorial, can you implement it this week? And after implementing it, does it save you measurable time or create measurable leverage?

If the answer to both is yes, the channel is worth your time. If you're watching for entertainment or inspiration but never building, you're consuming content, not scaling your business.

The founders who get the most value from these channels are the ones who watch less and build more. They pick one tutorial, implement it, test it, refine it, and move on. They're not subscribed to 15 channels. They're subscribed to the two or three that consistently teach workflows they can use.

That's the approach that works. Watch with intent. Build with focus. Refine until it's reliable.

What Happens When You Actually Build the System

The outcome isn't just saved time. It's leverage.

When you build an AI system that handles client onboarding, you're not just saving two hours per client. You're removing the bottleneck that kept you from onboarding more clients.

When you build a content repurposing workflow, you're not just saving time on social posts. You're creating compounding visibility without adding hours to your week.

When you build an email follow-up sequence that runs automatically, you're not just automating a task. You're making sure no lead falls through the cracks because you were too busy to reply.

That's the difference between using AI and scaling with AI. One saves you minutes. The other creates capacity you didn't have before.

Frequently Asked Questions

Who are the fastest-growing AI automation YouTube channels in 2026?

The fastest-growing AI automation YouTube channels in 2026 include Nate Herk, who went from zero to over 500,000 subscribers in under two years, Liam Ottley with 713,000 subscribers, and Mark Kashef, who has trained over 700 professionals on AI implementation. These channels focus on step-by-step tutorials that teach founders and professionals how to build AI workflows and agents that automate real business processes.

What makes these AI YouTube channels different from other AI content?

These channels teach implementation, not theory. They show you how to build the workflow on camera, step-by-step, so you can replicate it in your own business. The content is specific enough to implement but general enough to adapt. The audience isn't watching for inspiration. They're watching to learn how to build something this weekend that saves them time and creates leverage in their business.

What is the AI Automation Agency model that Liam Ottley teaches?

The AI Automation Agency model is a business model where agencies sell AI automation services to companies that need workflows built but don't have in-house expertise. The agency builds the agent or workflow, trains it, and delivers a system that runs without ongoing human intervention. Liam Ottley coined the term and teaches both the business side (client acquisition, pricing, project scoping) and the technical side (how to build the workflows clients are paying for).

What tools do these creators recommend for building AI workflows?

The most commonly recommended tools include n8n for workflow automation, Make for visual workflow building, Claude and ChatGPT for AI processing, and Airtable or Google Sheets for storing context and data. You'll also see ElevenLabs used for voice cloning and text-to-speech in video or podcast workflows, and Opus Clip for turning long videos into short-form social content. The tool choice depends on the specific workflow being built, but these are the core platforms that show up across most tutorials.

What is Context Training and why does it matter for AI automation?

Context Training is the process of teaching your AI everything it needs to know about your business before you ask it to do the work. Most AI systems produce generic outputs because they don't know your clients, your voice, your past work, or your decision-making process. Context Training solves that by giving the AI the specific information it needs to produce outputs you can actually use. You refine that context over time so the AI gets better, not just faster.

How do you implement what these channels teach without getting overwhelmed?

Start with one specific problem in your business. Don't watch tutorials at random. Search for the exact workflow you need to solve that problem. Watch the tutorial once all the way through, take notes, then build it. Test it with real data and refine it until it works reliably. Once one system is working, move to the next workflow. The founders who succeed watch less and build more. They focus on implementation, not consumption.

What's the difference between an AI agent and an AI employee?

An agent completes a task. An AI employee owns a role. A booking agent that finds one speaking opportunity is doing a task. A Speaker Booking Agent that pitches you to five stages every week, tracks replies, follows up, and owns your entire pipeline is an employee. The distinction matters because if you're still the bottleneck after implementing AI, you've only automated a step. The goal is to build systems that handle entire jobs from start to finish.

Why are founders watching these channels instead of the bigger AI names?

Founders don't need commentary or inspiration. They need instructions. The biggest AI YouTube channels are great for news and trends, but if you're trying to automate your client onboarding process, you need a step-by-step tutorial, not a recap of what happened at a tech conference. The channels growing fastest in 2026 are the ones that assume you're here to build something, not just learn about it. They solve the exact problem the viewer is stuck on right now.

Individual results vary. Time savings depend on your business, your tools, and how you manage your AI employees.