ContentPort

If you ever tried to create a blog post or a tweet with AI, you know that it always sounds very unnatural and robotic. And it's very difficult to get AI to sound like a real human. That's why people can immediately tell that something is AI generated which makes it unrelatable and spammy and they will probably just mute you and eventually the platform will shadow ban you because no one is interacting with your tweets. Today I'm going to show you a tool that uses AI to generate natural sounding posts for the X platform and it reads your own previous tweets to adopt your writing style. You can give it websites and files to read and you can actually schedule and queue your post right inside the app.

My name is Florian Walther and this is the AI tool corner where I review the latest AI software to find out which ones can actually improve our lives and businesses.

You can find the tool under contentport.io. The link is in the video description and it's free to use. Just go ahead and create an account. Then you have to connect and authorize your X account so that the app can read and post tweets on your behalf. Now the cool thing is it automatically reads your last 20 tweets that you posted on your account and it uses them to adopt your writing style so that the new posts it generates sound similar to your original ones. This is my coding account by the way.

Up here, we can fine-tune the writing style if we want. For example, I told it that whenever we post something funny like a meme, then it should write in lowercase because this just looks more natural. We are going to try this out later. You can do all this configuration in your account settings.

Now, before we try out the generator, let's also take a look at the knowledge base quickly because this is very cool. Here we can add different websites and also documents that the AI can later use to extract information and then it can write tweets based on the information in these documents. We can upload files or give it links to a website. For example, I have this React best practices course for my coding YouTube channel and I added the landing page to the knowledge base over here in the AI chat where we generate our posts later. We can then attach something from the knowledge base or attach a file directly from our computer. But we can actually also just post links into the chat directly. So let's try this out.

So here's a blog post I've written a while ago with tips on how to teach yourself coding. Build projects, read documentation, do a little bit every day and be consistent. We can actually just copy this link and paste it here in the chat. For example, we can say, “Please write a thread that summarizes this blog post.” Then I paste the link, enter. And now the AI goes ahead and reads this blog post and it generates a thread based on this information. So this looks really cool.

Now I just noticed that it wrote everything in lowercase, which is because of these writing style instructions that we gave it. Now I told it to only write in lowercase if it's something funny. And this post is not supposed to be funny, but it seems like the AI didn't really understand this. So maybe let's try improving this prompt. For all serious posts, write in normal sentence case. Let's save this and then let's try this again. I'm going to copy the same prompt in a new chat and let's see if the writing style is what we expect now.

Yeah, perfect. So it takes all these writing style preferences into account and now it generates a full Twitter thread based on this blog post that I gave it. And this is actually the writing style I have used myself in the past. If you take a look at my past tweets, I always write very short paragraphs, often just single sentences, and I always put an empty line between them to keep these tweets readable. And the AI adopted the same writing style. Very short paragraphs, empty lines between them and it actually sounds like me and it summarized this blog post with the information inside it.

Now if we don't quite like it, we can follow up in the chat. For example, we can say make it more concise and it will rewrite these posts in a more concise way. When we are happy with the result, we can use this tweet with this apply button. Now for this we have to go back to the studio which is where we actually design our post. Then we click apply and it gets added over here. Now we can modify it ourselves if we want.

Also they have a really cool image editor where we can upload and edit an image. Let's try this out. Doesn't this look cool? You can change the frame. You can change the design of the screenshot, rotate it, whatever. Change the background color. And then we can add it to our tweet. And when we are happy with the result, we can schedule our tweet and it will automatically post it for us. Let's say in half an hour schedule. And we can find and edit our whole schedule here. This is our tweet that's going to be posted at 10:30 a.m. And if we want, we can edit it, delete it, or post it right now. And we can schedule more posts afterwards. So, this is a fully functional Twitter scheduling system.

We can also schedule multiple posts at once. For example, I can attach my React best practices course and then tell it create a few funny memes to promote my React best practices course. Now again, it reads the landing page. And then it goes ahead and it creates multiple tweets at once. And remember, I told it that whenever I want to write something funny, it should use all lowercase because this sounds more natural, as if I just typed it casually. And it did exactly that.

Now, from my experience, AI in general is not very good at being funny because being funny is difficult for humans and especially for AI. So, I would probably mostly use it for serious posts and summaries and threads. And for this, it's great, especially if you can give it something to work with like a blog post that you have already written.

Lastly, let's take a look at the keyword monitor. Here we can give the tool a few texts that we want to observe. Those are programming terms. And then it automatically loads the most recent or most popular posts that contain these keywords. Right now we can't do anything special with it, but of course we can use this for inspiration. Or we could just copy paste this into our chat and tell the AI to generate something based on these existing tweets.

This tool is still quite new, but they keep adding new features and it works really well. And I know that they are planning to add additional platforms in the future. So it might go beyond just X posts in the future.

If you want to try out ContentPort, the link is in the video description below. Subscribe to the channel for more AI tool reviews in the future. Then I wish you a nice rest of the day. Take care.