AI Didn’t Make Knowledge Products Worthless. It Made Outcomes More Valuable.

These were supposed to be the years AI made knowledge products worthless. Why pay for a course when ChatGPT can explain almost anything? Why join a community when AI can answer your questions in seconds? Why buy another guide when you can generate something similar yourself in a few minutes?

On the surface, the argument made sense. Information got free. But something else happened at exactly the same time: guided outcomes got more valuable. And I think the market has already voted.

Information Isn't the Product Anymore

For years, a lot of online education was built around access to information. Pay for the course and get 40 videos. Join the membership and unlock 200 resources. Download the prompt library. Get the latest AI news. Access our template vault. The perceived value was often measured by how much stuff you got for your money.

That worked when good information was harder to find. Today, I can open an AI tool and ask it to create 100 marketing prompts before I've finished my coffee. I can ask it to explain Facebook Ads, outline a sales funnel, teach me the basics of SEO or create a 12-week marketing plan for a particular type of business.

That's why simply having more information behind a paywall isn't particularly valuable anymore. The internet already gave us too much information. AI has simply made finding, organising and understanding that information ridiculously easy.

Nobody Wakes Up Wanting Another Prompt Library

This is where I think a lot of course creators and community owners are getting things wrong. They're trying to compete with AI by producing even more stuff. More prompts, more resources, more templates, more videos and endless updates about the latest AI tool that launched yesterday and might be forgotten about next month.

But that's not what most customers actually want. The small business owner doesn't really want 500 marketing prompts. They want more customers. The consultant doesn't want another 20 lessons about LinkedIn. They want qualified leads. The business owner doesn't necessarily want to understand every new AI model released this month. They might simply want AI to save them five hours every week.

People aren't really buying information. They're buying the possibility of getting from where they are today to somewhere better. The information is simply one of the things required to help them get there.

The outcome is the product.

Build Courses Backwards

Instead of starting with the question, "What content should we add?", course creators should start with a much more useful question: "What should somebody be able to achieve after completing this?"

If the promised outcome is helping someone generate their first 10 qualified leads using AI, everything inside the product should point towards achieving that. You don't necessarily need 100 lessons, a daily AI news feed, a giant prompt library and another resource section that nobody will ever open.

You need a path. Show them where to start, what to do first, what comes next and what success looks like. Give them the templates they actually need, show them examples, get them to implement what they've learned and provide feedback when they get stuck.

That's where the value sits now. The value isn't simply in transferring information. It's in creating progression.

Communities Need to Think the Same Way

I've been around online communities for long enough to see the same mistake repeatedly. Community owners become obsessed with activity. How many posts were published this week? How many comments did we get? How many resources are sitting inside the library? How often should we post AI news to keep everyone engaged?

But a busy community isn't necessarily a valuable community. You could have hundreds of posts every week and still have members getting absolutely nowhere.

I'd rather have somebody join a community, follow a clear process and achieve something measurable than spend six months scrolling through an incredibly active feed. A good community shouldn't just give people somewhere else to consume content. It should help them move forward.

This is also where communities can provide something that AI alone struggles to replicate. Direction, accountability, feedback, context, experience and connection all become much more important when information itself is everywhere.

The question shouldn't be, "How do we keep members consuming?" It should be, "How do we keep members progressing?"

AI Didn't Kill Online Education

I actually think AI has the potential to make good online education considerably more valuable. What it has done is expose weak online education.

If your entire value proposition was access to information, then AI is a problem. If somebody was paying you £50 a month primarily to access prompts, articles and information they can now generate themselves within seconds, you're going to have to rethink what you're selling.

But if your product helps somebody take information and turn it into a result, AI can make that product significantly better. You can use AI to personalise learning, answer questions between sessions, simplify complicated subjects, provide feedback and help members implement what they're learning faster.

AI becomes part of the delivery mechanism rather than the thing you're actually selling.

That's an important distinction.

Stop Building Bigger Libraries

I think we're going to see fewer successful knowledge businesses competing on the size of their content libraries. Having 200 hours of training doesn't automatically make your course better than one containing two hours. Having 5,000 community resources doesn't automatically make your membership more valuable than one containing 20 carefully selected resources.

In fact, more content can sometimes make the product worse because you've simply given the customer another problem to solve: Where do I start?

That's probably one of the reasons they came to you in the first place. They don't need another mountain of information. They need someone to simplify it.

The strongest education products will increasingly remove decisions rather than create more of them. They'll understand where somebody is today, where they're trying to get to and what they need to do next. Everything else becomes secondary.

The Market Has Already Voted

People already have access to more free information than they could possibly consume in a lifetime. Now they also have AI capable of explaining that information, summarising it and tailoring it specifically to them.

Yet people continue paying coaches. They continue joining communities. They continue buying courses. They continue hiring consultants.

That's because knowing and doing are completely different things.

AI has dramatically reduced the cost of knowing something. It hasn't eliminated the difficulty of actually doing it, sticking with it, knowing what matters, getting feedback when something goes wrong and continuing until you achieve the result.

That's the opportunity for anyone building a knowledge product today.

If you're running an online course, membership or community, I'd worry less about how many prompts you can give people. I'd worry less about publishing every piece of AI news. I'd worry less about adding another 50 resources because you think the library needs to look bigger.

Instead, ask one much simpler question:

What result is somebody paying us to achieve?

Then build the course, community, support, content and experience around helping them achieve it.

Information got free.

Outcomes didn't.

And that might make genuinely useful knowledge products more valuable in the AI era, not less.