A Seven-Step Workflow for AI-Assisted Publishing (Full Transcript)
[00:00:00] Speaker 1: going to show you how I was able to publish my article just from my voice to the publication. I'm going to give you a step-by-step. There are seven steps that you can follow to be able to do that. At the end of the day, you'll be able to publish your article, especially when you want to get your ideas out there and you have limited time and resources. This is the opportunity for you to do that, especially in the space of AI, using AI to help you to make your ideas known, to publish your information for people to read and learn about it. This is a great opportunity for you. And I'm going to show you step-by-step from the beginning to the end. It starts from your voice. But before I give you the steps, there's one thing that you have to do. You have to be knowledgeable about the subject matter. My article was about disclosure, the need to show your work when you use AI. In order to start the article, I read a lot about it. I've been teaching about this, the need to disclose your work. So I am familiar with the subject matter. So the first thing that you have to do before even you start thinking about writing is to read and learn more about a topic. Then after learning more about the topic, what do you have to do? You have to develop a framework that will help you to communicate what exactly you want to share to your audience. So after learning more about the subject matter, what do you have to do next? The first step is to dictate. What is dictation? It's all about putting your ideas on paper. So when you're using technology, what do you have to do is you have to go to Word document, and then you click on dictate, and then start talking about what you want to share, right? So you talk about it. That's what I did. I just keep on talking and the system will dictate for for about two pages. And then after that, I review what the system has given me, and then make the necessary corrections. After making the necessary correction, I went to the stage of review. That's the second stage. I uploaded the document into the system, and I asked, I'm working on this topic, and I provided what I'm talking about. So you have to give the system a little bit of context so that the system will know what exactly you're doing. So you say that you are writing about this topic. Can you critically review and provide me a feedback so that it will help me to improve? Also, can you suggest sources that can be connected to the ideas that I have? So the system will go through all the information that you have provided, and then provide you critical feedback, and also sources that you can use to improve as you're trying to work on your article. So that's what I did. I got all the feedback from these three sources, and then I reviewed all of them. Sometimes what I did was that I accepted some of the suggestions. Sometimes I rejected the suggestions. And so the most important thing here is that you are writing the paper. See all this information that you get from AI as suggestions. You have to review and accept. If it makes sense to you, you accept. If it doesn't make sense, you reject, right? That's what I did. So after that, I came up with five headings, and then what I did was that I opened the Word document, and I went to add int. I click on add int, and then what I did was that I look for Cloud AI, right? So if you have access to Cloud, you can use that. You can go to, I look for Cloud, and then I click on it to bring that information so that I can use the Cloud on the Word document. So I selected Cloud, and then I start chatting with Cloud concerning my document. So what I did was that I told Cloud, I have five headings. Can you review my current document and then categorize them into these five headings? And the system was able to do it within seconds. The system was able to go through my document and then group them based on the five topics I've given to it. So then that's the test, which is structure. Cloud helped me to structure. You always have to review. I make sure that is that division or categorizing makes sense? If that grouping doesn't make sense, you have the final say. You are writing the paper, not AI. AI is just a tool supporting your writing. So then I move to the fourth thing, which is refine. So this is the stage where I go heading by heading, read everything, reread it, making sure that it's making sense, and then also incorporating suggestions that the system has provided to me in terms of references. So all the references the system has provided, I read and also check the references, check the articles, and then make sure that the information that is linking to my ideas are the same or are similar. Then I ask Cloud to review what I've done and then ask as an editor. So technically what I did was I asked Cloud to edit my document for me and also use track changes so that I will be able to go through and review, accept, and reject. So I always have the option to accept or reject. So Cloud went through my document, making sure that the citations are correct, making sure that the grammar is good, is corollary written, and also one thing that I also did is that you can also do the same. If you have already published an article or you have a document that you wrote, you can upload it into the system so that the system will learn how you write, so that the system provides you feedback based on how you write. Jenny AI is AI research and academic writing assistant. So imagine you have an AI platform that is solely specialized in writing, helping you to write in an academic way. This tool will be the best for you. See what I've provided to you all these steps. Imagine you have a platform that you can do almost all of the steps that I've provided to you in one place. Then this software will be the best. If you have a draft, you can go to Jenny AI and then upload the draft. You can chat with your document and then the system can critique your work. You can ask the system to review as an editor and based on track changes you can accept or reject. If you even don't have a draft, you can start with your research ideas. You can start a project and then the system will give you a format and you can start typing. As you are typing, the system will also suggest ideas that you can review and incorporate in your writing. And you can also import your research articles. You can use this one to set relevant articles that can be connected to your study. So it's like all the steps that I've provided to you in terms of writing an article, you can do most of the actions here in one place. From the drafting to an article or manuscript ready for publication. You can try it out for free and let me hear from you about your experience. Sign up for Jenny and use this promo code EDU20 to get 20% off. You can also use the link in my description to go to the website. Thank you Jenny AI for sponsoring this video. The fifth stage is called verify. So this is where as a researcher or as an author, I went through all the documents again making sure that all the citations are right. So I review and make the necessary corrections when needed. I make sure that all the citations are accurate. So the next step is to disclose. So this is a thing that I really did. As I was going through working on the document, I also was keeping record of the actions and decisions that I'm making and also what the role AI is playing in the process. So this is what I did. I went to ChachiPT. I told that I'm writing an article and I want to document the actions that I'm taking as I use AI to help me to work on it. So any decision that I make, I go to ChachiPT and then document that information. So normally what I do is that I go to ChachiPT, I click on the microphone and just start talking. I told ChachiPT the first step, the second step, all the steps and the role AI took in that process. So the system was documenting all the actions that I'm taking. So after that, I asked ChachiPT to summarize all the actions that I took. I review it and then I was able to write the disclosure statement. So the disclosure statement, let me show you how it looks like. So as you can see here, I have my disclosure statement here. It details the main actions that I took. You don't have to provide every information. This is just the main information that you can provide, showing people the role AI played in the process as you are working on the document and also the actions and the decision that you took. So this is what I provided. So that's the C's point. The C's point is disclosure. You have to disclose. And then you can use this framework to disclose. So the framework is called ROAD. R-O-A-D. R stands for request. What kind of questions do you ask AI? And O stands for output. What kind of output did you receive? And what did you do with the output? This is called assess. How do you assess the output? And then what kind of decision did you make? Did you accept, reject, or refine the information that you got from the AI? So you can use this framework to write your disclosure so that people will believe what you found. And this is also going to help other researchers to learn about the strategies that you use to produce an article. So what are the strategies? And it also opens up for people to give you feedback, to critique your work. And as they critique, you learn from their suggestions and then you improve the way that you use AI. So that's the best way to do that. You have to document and you have to report how you use AI to do your study or to write an article. So the last one is share. So the sharing is more about publishing. So what I did is I went to this website. I went to znodo.org. So this is the website that you can publish your article. It's not a peer review article. It's an article that you want to share your ideas without going through the process of peer review. Sometimes you want to quickly share your ideas like what I have here. This one doesn't replace peer review article. This is another avenue to quickly share your ideas so that people will get to know and learn from. So you can upload your article here. It gives you the DOI and then you'll be able to share and people can read and download. There's 22 downloads as you can see here. So these are the steps that you can follow to create an article. As you know, start with the TATE. I'm not saying that this is the only strategy but one of the strategies is to just document what is in your mind by talking about it and having a system to record and also transcribe what you are saying. So the second step is to review. So after the TATE, you have to review the transcript making sure that everything is right. And the next one is structure. You have to put the ideas in a specific structure. But before you do that, you can also use AI to critique your work and provide you feedback and resources that you can use to improve. And then you go ahead and structure based on specific headings that you want. And then you can also use AI to help you to refine your document by providing you feedback as an editor. And then after that, you also have to review and verify whether everything is right. And you can disclose the AI use. And then lastly, you can share for the world to know about the idea that you want to share. So if you want to have access to the article, I'll put a link in the description. You'll be able to find that. Let me hear from you. What are your thoughts? This is the strategy that I use. Let me hear what are additional strategies that you use to write articles in terms of incorporating AI into the writing. Let's learn from each other. Thank you so much for your time.
- Build subject-matter knowledge and a communication framework before drafting.
- Use dictation to capture ideas quickly, then correct the transcript yourself.
- Ask AI for critique, improvement suggestions, and relevant sources, but treat every response as a suggestion.
- Structure the draft around clear headings and use AI editorial help with track changes for reviewable edits.
- Verify claims and citations directly; the author is accountable for accuracy.
- Document AI involvement and disclose it using the ROAD framework: Request, Output, Assess, Decision.
- Share non-peer-reviewed work on repositories such as Zenodo for rapid public access, without treating it as a replacement for peer review.