Errors with ChatGPT Integration for Column Creation and Enrichment

Drake T. · 2024-10-16 19:15:29.560983+00

I’m getting some errors when using the ChatGPT integration. No matter what prompt I use or API key I select, creating columns and starting the enrichment fails.

Errors with ChatGPT Integration for Column Creation and Enrichment | Clay

12 comments

· Sorted by Oldest

Channeled
APP
·
·
👋 Hey there! Our support team has got your message - we'll be back in touch within 24 hours (often sooner!). If you haven't already, please include the URL of your table in this thread so that we can help you as quickly as possible!
Drake T.
·
·
The error only appears to affect the tab leveled Contacts - Master 1 on the linked workbook. Re-running existing integrations appears to work and adding a new ChatGPT column works in the tab labeled Contacts - Master 2.
Drake T.
·
·
I also know it’s not a column limit error, as the tab only has 35 columns.
Bruno R.
·
·
Hi Drake, thanks for reaching out. Taking a look at this now.
Bruno R.
·
·
It seems the issue stems from missing inputs. You're referencing a column containing post content, but most rows have empty values, which prevents GPT from processing them effectively. I’ve recorded a video explaining this in more detail: https://www.loom.com/share/12cc2014fa864694ab5c9bde46029963?sid=f998cb7b-f984-4f3c-b084-7125db1b94e8
Drake T.
·
·
Hey Bruno R. thanks for the quick response! Unfortunately, that’s not the issue I’m having. The columns you showed in the video are operating exactly as I want them to. The issue is with creating a new column with the ChatGPT integration.
Bruno R.
·
·
Hey thanks for adding context here and sorry for missing the GIF you recorded - this is super helpful! I was not able to reproduce the issue on my end. Do you mind sending over a video of the step-by-step process you went through to run into the error toast messages I see in your interface? If you're still experiencing this issue, can you kindly perform a hard refresh for me? Here are some instructions on how to do it: Mac: Chrome and/or Firefox = Press Cmd + Shift + R Safari = Press Cmd + Option + RPC:Chrome, Firefox, Microsoft Edge = Press Ctrl + F5
Drake T.
·
·
It seems like the error has something to do with the specific prompt I’m using. If I try to use a specific prompt from my library, I’ll get the error I just showed. However, If I create a ChatGPT column with just the word “test” as the prompt, it’ll create the column and start generation no problem. I can then edit the column and change the prompt to the one I want to use from my template library and it works just fine. Here’s a video for demonstration: https://s.demand.inc/04unvYj1
Drake T.
·
·
Bruno R.
Bruno R.
·
·
It looks like we've got a bug on our hands - I've escalated this over to our engineering team and we'll let you know as soon as we know more!
Bruno R.
·
·
Good news! Happy to report that we heard back from the team and this issue should be fixed. Thank you for your patience as we worked on it. Please let us know if the problem has been resolved on your end, or if you still have trouble.
Channeled
APP
·
·
Hi Drake T.! This thread was recently closed by our Support team. If you have a moment, please share your feedback:

Channeled
APP
·
·
👋 Hey there! Our support team has got your message - we'll be back in touch within 24 hours (often sooner!). If you haven't already, please include the URL of your table in this thread so that we can help you as quickly as possible!
Drake T.
·
·
The error only appears to affect the tab leveled Contacts - Master 1 on the linked workbook. Re-running existing integrations appears to work and adding a new ChatGPT column works in the tab labeled Contacts - Master 2.
Drake T.
·
·
I also know it’s not a column limit error, as the tab only has 35 columns.
Bruno R.
·
·
Hi Drake, thanks for reaching out. Taking a look at this now.
Bruno R.
·
·
It seems the issue stems from missing inputs. You're referencing a column containing post content, but most rows have empty values, which prevents GPT from processing them effectively. I’ve recorded a video explaining this in more detail: https://www.loom.com/share/12cc2014fa864694ab5c9bde46029963?sid=f998cb7b-f984-4f3c-b084-7125db1b94e8
Drake T.
·
·
Hey Bruno R. thanks for the quick response! Unfortunately, that’s not the issue I’m having. The columns you showed in the video are operating exactly as I want them to. The issue is with creating a new column with the ChatGPT integration.
Bruno R.
·
·
Hey thanks for adding context here and sorry for missing the GIF you recorded - this is super helpful! I was not able to reproduce the issue on my end. Do you mind sending over a video of the step-by-step process you went through to run into the error toast messages I see in your interface? If you're still experiencing this issue, can you kindly perform a hard refresh for me? Here are some instructions on how to do it: Mac: Chrome and/or Firefox = Press Cmd + Shift + R Safari = Press Cmd + Option + RPC:Chrome, Firefox, Microsoft Edge = Press Ctrl + F5
Drake T.
·
·
It seems like the error has something to do with the specific prompt I’m using. If I try to use a specific prompt from my library, I’ll get the error I just showed. However, If I create a ChatGPT column with just the word “test” as the prompt, it’ll create the column and start generation no problem. I can then edit the column and change the prompt to the one I want to use from my template library and it works just fine. Here’s a video for demonstration: https://s.demand.inc/04unvYj1
Drake T.
·
·
Bruno R.
Bruno R.
·
·
It looks like we've got a bug on our hands - I've escalated this over to our engineering team and we'll let you know as soon as we know more!
Bruno R.
·
·
Good news! Happy to report that we heard back from the team and this issue should be fixed. Thank you for your patience as we worked on it. Please let us know if the problem has been resolved on your end, or if you still have trouble.
Channeled
APP
·
·
Hi Drake T.! This thread was recently closed by our Support team. If you have a moment, please share your feedback: