Save the date! Join us on October 16 for our Product Ops launch event. Register here.
Apr 27, 2023 09:32 AM
I'm trying to use ChatGPT to analyze written data from airtable surveys, but ChatGPT is unable to access the information from a csv file. Is it possible to connect ChatGPT to Airtable in order to give it access to specific tables? The idea would be to be able to ask ChatGPT questions such as "What are the top 3 things guests enjoy about this trip?" based on the survey responses.
Solved! Go to Solution.
Jan 12, 2024 05:44 AM
I have recently written a guide for doing something like this:
https://www.csvgetter.com/blog/enable-airtable-ai-analysis-with-perplexity
Apr 27, 2023 08:13 PM
Apr 28, 2023 03:56 AM - edited Apr 28, 2023 03:57 AM
Yes, you can very easily integrate & automate ChatGPT for free with Airtable by using Make’s ChatGPT module.
I also show off how to do this on this episode of the BuiltOnAir podcast.
And we are having tons of ChatGPT and Airtable discussions here.
Apr 28, 2023 12:38 PM - edited Apr 28, 2023 12:39 PM
> ask ChatGPT questions such as "What are the top 3 things guests enjoy about this trip?" based on the survey responses.
I don't get the sense that either of the responses (from @ScottWorld or @Lom_Labs) addresses this requirement using GPT. I would love to see how [exactly] GPT can give you this outcome.
Apr 28, 2023 05:01 PM - edited Apr 28, 2023 05:02 PM
You would have to aggregate records together and then send the data to ChatGPT to analyze. For example, Make has several "aggregator tools", which can aggregate data from multiple Airtable records together into various outputs (i.e. text, tables, arrays, JSON, etc.). Make's documentation is absolutely terrible, but they have a Text Aggregator Tool, a Table Aggregator Tool, an Array Aggregator Tool, a Number Aggregator Tool, JSON creation tools, and more.
However, ChatGPT has a limit on how much data you can send it in a request, so that would limit the amount of data you can aggregate together in Airtable to send to ChatGPT.
Apr 29, 2023 09:55 PM
Apr 29, 2023 10:51 PM
Hello @Bill_French !
It is easy to do this by first creating a summary table that can aggregate and consolidate the survey responses. This can be accomplished by utilizing rollup fields, which allow us to group and merge data from multiple records from the survey response table.
Once the survey responses have been merged and consolidated in the summary table, a formula field can be added to concatenate a prompt to the consolidated responses. This will enhance the clarity and cohesiveness of the final message that is to be sent to ChatGPT.
Finally, an extension can be utilized to send the combined and enhanced data to ChatGPT!
Apr 30, 2023 06:55 AM
My hatred for Khoros continues. I had a really nice post with code and everything. It was lost when I accidentally changed views.
So, I recreated it here. Enjoy...
May 02, 2023 11:21 AM
After reading through everything (particularly @Bill_French 's fascinating article), it seems as though what I'm looking for - essentially a direct connection between AirTable and ChatGPT, where ChatGPT can access the information in AirTable on command, is not an option. The solutions proposed are a bit over my level at the moment but if I'm able to figure it out, I'll post an update here!
May 02, 2023 07:01 PM
> ... essentially a direct connection between AirTable and ChatGPT, where ChatGPT can access the information in AirTable on command, is not an option.
If you want your data users to access Airtable data directly from the ChatGPT user interface, you must build a ChatGPT Plugin. This complex approach requires security and API code that will make it seamless for your users to access their private data from OpenAI's UI products. I have my doubts that you actually want this; here's why.
As I explained in the article, if you want your users to be able to use their data with ChatGPT (not to be confused with inside ChatGPT), you need to build an extension app like I demonstrated, that blends your table data with AI to create a UI that allows users to query the data through natural language. Example query...