It depends on how the AI image generator sends back the final image to you.

1. If the AI image generator sends the image back to you with a public URL where you can access the image, then you can just use Make’s Airtable modules to insert the URL into an Airtable attachment field. Airtable will then download the image into the attachment field.
    
2. If the AI image generator sends the image back to you as a file, you will need to upload the file to a cloud storage space — I like using the Google Drive “Upload A File” module, which will return a URL for you that you can send to an attachment field in Airtable.

Hope this helps! If you’d like to hire the best Airtable consultant to help you with anything Airtable-related, please feel free to contact me through my website: Airtable consultant — ScottWorld

Appreciate it, thanks! 

Hey ​@metricsmule what type of images are you looking to generate?

You might be able to achieve all of this by leveraging Airtable AI features. Check out this article on image Generation. 
 

Mike, Consultant @ Automatic Nation

Thanks ​@Mike_AutomaticN 

I have a YouTube channel that focuses on AI, and more specifically generating AI images and videos. My favorite tools are Freepik, Leonardo AI and Midjourney.

So I’d love to have an automation like this: 

1. Enter basic prompt into Airtable
2. Routes to Make/Zapier
3. Connects to ChatGPT -- ChatGPT then take one of my prompt improving generators and uses that to 10X my basic prompt in Airtable. 
4. Bonus: ChatGPT sends to an AI image app, which then generates the improved image/prompt. 

Hey ​@metricsmule, another option beyond Make/Zapier is Data Fetcher. It’s an Airtable extension so it works right inside Airtable.

Tutorial: Create AI-Generated Images in Airtable with OpenAI & DALL·E

Explore all Airtable integrations

(Full disclosure, I’m the founder)

Hey ​@metricsmule!

You can probably set something similar up within Airtable itself having:
1.  Long text field for manual prompt
2. AI field for AI enhanced prompt
3. AI image generation field from enhanced prompt.

However, the above would still use OpenAI.

If you would want to use a different model/service for the image generation, you can absolutely set up an automation using pretty much any automation tool such as Zapier, Make, or N8N (I would personally suggest n8n -more info here).

Sorry for the delay, but if you are still needing this I’d be happy to help you out. Just feel free to grab a slot using this link.

Mike, Consultant @ Automatic Nation