Thank you for contacting me. Please note that I live in the GMT+3 time zone - responses might be delayed by this.
This topic has 8 replies, 2 voices, and was last updated 1 month, 3 weeks ago by 
.
-
Posts
-
-
Hello,
We have recently started using Aiomatic’s Bulk AI Post Creator, but we are facing a few critical issues that make it difficult for us to use the plugin in a fully automated publishing workflow:
Prompt Ignoring Issue
Sometimes the plugin does not process the prompt we provide.
Instead of rewriting or generating the article according to the custom rules in our prompt, it simply outputs generic ChatGPT responses such as “provide a text to be rewritten” or similar.
This behavior wastes tokens and also results in unusable content being automatically published on our site.
Since our intended workflow is automatic publishing without manual checks, this inconsistency makes it very risky to rely on Bulk Creator.
Prompt Rules Not Always Applied
Even when the plugin uses the prompt, the generated articles sometimes ignore critical rules we defined (e.g., preserving all specific factual details, keeping exact numbers/dates, including all quotations fully).
This forces us to double-check outputs manually, which defeats the purpose of automated publishing.
Could you please clarify:
Is this a known issue with Bulk Creator prompt handling?
Are there settings we should adjust to ensure the plugin always follows the given prompt without skipping it or defaulting to a generic ChatGPT message?
Is there a way to reduce token wastage in such cases?
Topic-Based vs Title-Based
Could you also explain the exact difference between topic-based and title-based modes in Bulk Creator?
From our understanding, topic-based seems to generate posts around a general subject, while title-based generates posts directly tied to a provided headline. But we would like confirmation on how the plugin handles these two modes internally.
Since our main goal is to run Bulk AI Post Creator in fully automated publishing mode with no manual review, we need to be sure that prompts are always respected and outputs are reliable.
Looking forward to your guidance.
Best regards,
Hazel -
Hello,
First of all, thank you for your purchase.
This is an unusual issue, I never got a similar issue reported before. Which AI model are you using in the plugin? Note that some older models might have limited context window, making their understanding be limited. I recommend you try using a newer model, like gpt-5, gpt-4.5 or gpt-4.1. Experimenting with the o1 and o3 models might be also an option.
Regarding your other question, the ‘Title based posting’ will use a single prompt and instruct the AI to write the article in a single shot. This can work well, but will provide limited length articles, as the AI models can provide only limited length articles at the moment.
The ‘topic based posting’ will ask a post topic and call the AI multiple times, to create an outline of the article in the first phase and afterwards, will call the AI to create each chapter of the article separately. This will allow much longer articles to be created, but might create issues in the article’s context, if the prompts and AI models are not set up the right way.To get 100% accurate articles, I recommend you try using only the title based posting.
Regards,
Szabi – CodeRevolutionRegards,
Szabi – CodeRevolution. -
Hello Szabi,
Thank you for the quick response and detailed explanation.
We are currently using GPT-5 in the plugin, but unfortunately the issue still persists — sometimes the Bulk AI Post Creator does not process our prompt correctly and returns ChatGPT-like responses (e.g., “provide a text to be rewritten”) instead of generating the news article according to the provided instructions.
Since this happens inconsistently, it’s quite difficult for us to identify the root cause on our side. Would it be possible to schedule a short screen-sharing or troubleshooting session so you can see the behavior directly? That would help clarify whether it’s model-related or a plugin handling issue.
Regarding the posting mode:
We mainly generate news content (not general articles or SEO-style posts). Each post needs to follow a strict news structure and factual accuracy, so context consistency is very important for us.
Given this, could you please advise which option — Topic-based or Title-based — would be more reliable for short, factual, and time-sensitive news rewrites?Thanks again for your time and support — we really appreciate your help optimizing this workflow.
Best regards,
Hazel -
Hello,
Can you send me, please, temporary admin login credentials to your WordPress install, so I can check this issue out? Send it, please, to my email address: kisded@yahoo.com
Regards,
Szabi – CodeRevolution. -
-
Hello,
I checked also spam, but I did not get the credentials for your site, please resend it manually to my email kisded@yahoo.com
Regards,
Szabi – CodeRevolution. -
-
Hello,
I checked also spam, but I don’t find the email from your site, please resend it manually to my email address: kisded@yahoo.com
I suspect the previous email was not sent correctly.
Regards,
Szabi – CodeRevolution. -
You can try also alternative email address: support@coderevolution.ro
-
The topic ‘Bulk AI Post Creator Prompt Handling & Clarification on Topic vs Title Based’ is closed to new replies.