Hello everyone,
I’m a graduate student from Taiwan, currently conducting an educational research study on the effectiveness of conversational AI in improving high school students’ English language skills. For this project, I’ve created a Custom GPT on OpenAI, which my students (35 per class) are using to engage with English learning exercises.
However, during our first session today, I encountered an unexpected issue. About 5-6 students quickly received a message stating that they had hit the usage limit after asking 1~5 questions and were prompted to upgrade to ChatGPT Plus. This was surprising because it’s essential for my study that all students can interact with the AI without interruptions, at least for the month that I’m collecting data.
I would like to know if anyone else has experienced similar limitations with Custom GPTs used in group settings or if there’s a way to enable higher usage limits for educational purposes. What language model does Custom GPT use? Does that allow my students to have unlimited access to the Custom GPTs I designed if I turn off the web-browsing and DALLE3-generation capabilities of my Custom GPTs?Is there an option to pay for an upgrade to allow consistent and stable access for my students? Any advice or insights would be greatly appreciated! Thank you
ChatGPT is not meant for children under 13, and we require that children ages 13 to 18 obtain parental consent before using ChatGPT. While we have taken measures to limit generations of undesirable content, ChatGPT may produce output that is not appropriate for all audiences or all ages and educators should be mindful of that while using it with students or in classroom contexts.
We advise caution with exposure to kids, even those who meet our age requirements, and if you are using ChatGPT in the education context for children under 13, the actual interaction with ChatGPT must be conducted by an adult.
You also should not actually be using ChatGPT, which is an advertising platform to create an account and upgrade for further use of a GPT or to create your own customizations, and other adult activities such as paying for monthly subscriptions. You are relying on a commercial service that has no guarantee of allowing any free trials of the GPT created in a ChatGPT Plus account.
It is much better to develop on the API. There, you will be centrally-billed for actual usage, and you can develop a customer or student-facing application with your own levels of accounting and monitoring.
@morrismorris123123
I’m doing something similar for my students and realised my customGPT hits the limits fast also.
Did you find a work around?
I’ve looked at plugging in an API but the latency would just kill the entire project …
How so?
The API should have the same - if not better latency than ChatGPT itself.
Are you talking about the OpenAI API?
You might be integrating it in a different way that is suboptimal, however the API would be a perfect solution to this issue - without additional latency.
I believe the API will kill the latency because I am heavily using the voice chat, the whole custom GPT is like a conversation.
Within the CustomGPT it is all automatic and internal or native.
With API I believe I´d need server functions, or set up some server environment via Python. Let me know if you think there is another way.
Thanks! @j.wischnat I had a look at the link and I think I’d need a developer to implement it (like you said).
At the moment I just wanted to use the custom gpt and test that my prompts are valuable for my students.