If you are using content to market yourself or your business online, you're fundamentally trying to do two things: spread your POV and develop trust with your audience.

It’s harder than ever to find distribution, so it’s easy to look at outsourcing your content creation to AI, but done incorrectly, this tactic can backfire.

Overreliance on AI to create your content will dilute your POV, make you sound like everyone else, and worst of all, risk breaking trust with your audience. However, there are ways to use AI to 3x your content creation without devolving into AI slop.

The bad news is you are still needed in the process.

Here are 5 ways I use AI to scale my content:

AI for customer research

Using AI for customer research isn’t groundbreaking, and I think customer research for product development is a great early use case for many marketers. Using customer research to build out your content engine is a different process, though.

When I feel stuck, I answer potential customers’ questions, and those are constantly changing. There are two methods here.

You can use AI to dive into your customer tickets in Zendesk, but those questions tend to be more down-funnel, so the other way is to have AI search for questions for your prospects. The best places to have AI search are on Reddit and X.

There are different ways to do this, but I would suggest a few things:

• Make sure the AI is focused on recent activity and comments so you can find areas that are emerging where you can inject your POV into the conversation.
• Have the AI pull actual questions with the actual language where it can.
• Have it prioritize 2-3 questions, but also keep a list so it doesn’t repeat them over and over.

The reason I suggest 2-3 questions at a time is that you don’t need a backlog. While there will be evergreen questions, you need to spread your POV, and being part of a current conversation goes a long way.

Transcribe content instead of writing it

If you do written content, you need to try speaking your content and having AI transcribe it. This is considerably faster than typing, and I start most posts out this way.

I want you to avoid sounding like everyone else, so when doing this, you want to ask your AI assistant to keep as many of your phrases as possible as they take the messy transcript and turn it into something usable.

This is a quick jumpstart to what is usually a solid first draft. Yes, you need to edit it, but I have found this can cut writing time down by 50%.

The best part is the AI doesn’t have to try and “get your voice.” It just needs to take your actual voice and put it on paper.

I experimented with having AI write first drafts without this step, and I was constantly disappointed in the results. While it can take a topic and make some solid points, it always feels devoid of life.

I want my content to have my voice, my stories, and of course my bad jokes. Transcription makes this possible.

Repurposed content

So far, all of those were heavily human in the loop. With this next one, you can begin to dip your toes into autonomous AI creation.

When I talk about repurposed content, there are a few ways to do this, but most often this is taking long-form content and turning it into short-form content. Every blog post can become a LinkedIn post. Every YouTube video can turn into 10 TikTok videos.

This can be hands-off because AI doesn’t need to create anything. It is just taking what you already wrote or said and cutting it down into smaller content.

The bigger the backlog of content you have, the easier it is to do this one.

This is an area I am actively developing more structure around, so if you want to learn more about the system that allows this to work, make sure to subscribe.

This is an important one because we get bored of our message long before anyone else does. Most people, even those who subscribe to you, don’t see most of your content, so you need to repeat yourself. AI can do the repetition and allow you to continue to build the net new.

Over time, my goal and my suggestion would be to have AI repurposing 70-80% of your content from content you’ve already created. This will take time to build the catalog and system, but it will be rewarding when it’s humming.

One final thing on this one: you also don’t need to stay in the format you’re in. You can use video content to create text content. You can use blog posts to create image carousels.

Use AI to edit your videos

I’ll be honest, this one is still new for me, and I’m currently learning more about this one. The recent AI models from Claude and OpenAI have really upped their game when it comes to AI editing.

Think about this: you can create a yapping video in a few minutes, but even those of us who are comfortable speaking on camera will need to edit the videos. You need to add an on-screen hook. You need to add captions. You need to cut out dead air and remove filler words.

All that takes time.

AI can do all that for you while you record your next video.

In my early tests here, I see a lot of promise. The agent used my brand colors, used multiple text overlays to match the spoken content, and it removed extraneous speech to create a tighter video.

As I said, I am still working on this one, but I have been impressed that it can watch the video and come up with suggestions for cutting or even rearranging aspects of a video. I created a video recently, and AI cut it down from three minutes to a bit over one.

This warrants more testing for me, as it cut the videos to make them punchier, but it also removed some of the stories I told. This may be me enjoying my own ego, but I think those elements will develop trust over time, so over-optimizing for punchy may not be the best thing.

Build your context system

All of these suggestions have focused on how to amplify your voice and keep yourself at the center, but what if you want to push the limits? If you want to have AI creating content on its own, you need to build a context system.

This system needs to know everything about your product, who your customers are, what they are about, and more. This system needs your content. And most of all, this system needs constant updating to stay current.

The good news is you really want to build this system anyway.

I use an AI chief of staff who helps me organize my days, helps me follow up on tasks, helps me focus on jobs that move the needle while also picking up projects where they can.

The same principles in building the COS can be used to create content. This system turns you into an approver instead of a creator.

While this system has the highest volume potential, it has the most risk. You are avoiding sounding like everyone else by having a strong POV and customer understanding, but you still have the risk that comes with misleading people.

My one rule of AI is this: if you wouldn’t want someone to know you used AI for a task, don’t use it for that task.

However, I do think with the right system underneath, there is a lot of potential for this one to create unique content at scale.

Human-in-the-loop content is still the expectation

Every suggestion here is focused on how to use AI to help amplify what you are building without outsourcing all your content to an AI agent. You can search up prompts to do this work fully for you, but what you actually want to do is build systems.

Prompts may be able to find you quick wins, but I want you to build a system where you can create consistent distribution. If you are marketing a product or company, you want to build trust, so virality isn’t the goal.

That is why these ideas are developed around keeping a human in the loop.

You are what makes your content unique. You give your content life. It’s the same reason that, if you post the same content on your LinkedIn and your company’s LinkedIn, your personal one will almost always perform better.

With that said, content is also a volume game, so using AI is no longer optional. It’s an important part of creating scale.