I have a chatbot that’s integrated with an OpenAI assistant via API, running on n8n. The chatbot has access to some functions like checking schedules and booking appointments. Occasionally, it gives strange or inconsistent responses. For example:
Bot’s Response:
vbnet
Copiar código
For tomorrow, 11/13/2024 (Wednesday), we don't have any available times in the morning. The available times are:
- 1:30 PM
- 2:30 PM
Please let me know if any of these times work for your husband.
I will check the availability for tomorrow morning, 11/13/2024 (Wednesday). One moment, please.
As you can see, the bot first states there are no available times in the morning, but then says it will check for morning availability.
Ensure to copy your n8n workflow and paste it in the code block, that is in between the pairs of triple backticks, which also could be achieved by clicking </> (preformatted text) in the editor and pasting in your workflow.
```
<your workflow>
```
That implies to any JSON output you would like to share with us.
Make sure that you have removed any sensitive information from your workflow and include dummy or pinned data with it!
You might need to update the prompt and ask AI not to check the same day if it is already checked the availability for that date. The more detailed instruction you provide the likely AI will do what you expect of it.
I just joined, and I have been working with ChatGPT and OpenAI from the start. I’m a Web Designer & Developer of nearly twenty years. AI is either super advanced and is messing with us, or it is just too inconsistent to ever truly be helpful. I’m at the point where I’ve realized that AI might be a very, very large scam on all of us. They can’t get it to work, because they don’t know how it works. Lol, I’ve fixed my bot 10 times, and it’s like I’m unable to get it to output anything consistent. AI is a huge scam. This is just within the last hour, and I have dozens of these stories. Early on I thought I was tripping, but it turns out that AI will intentionally mess with you too. Lol, I wish I was lying. It is too unpredictable to ever be more helpful than chatting with it. Seriously, I mean think about it, everything should be automated and technology would be advancing at break-neck speed already if AI always followed directives, lol, I would already have a whole work-force if it was consistent.