Q:
I’m using the "Extract Text (2 tasks)" action to pull some text, but I keep getting this error:"An error occurred. The content was judged to be potentially harmful, so it failed. Please modify the input and try again."
How can I avoid this error?
A:
This error happens when the AI thinks the text might go against its content guidelines.
But don’t worry, there’s a simple fix! You can usually avoid the error by making a small change to the text you’re trying to extract.
📍 Here’s an example:
Let’s say you want to extract parts of an address like "1234 King Street, Downtown, Toronto, Canada," and you want to grab things like "Street," "Area," "City," and "Country." The AI might flag this as harmful, even though it’s just an address.
☝️ What You Can Do:
When you are trying to extract only address-related information, the AI might flag it. In this case, try adding a small adjustment to the text string, such as inserting the keyword "Address:" at the beginning.
For example, change it from:
"1234 King Street, Downtown, Toronto, Canada"
To:
"Address: 1234 King Street, Downtown, Toronto, Canada"
This small change can help you avoid the error and get the data you need!
📚 Related Guides
How to set up the "Data Extraction" action using AI - View Guide Here
How to Use AI for Data Transformation - View Guide Here
🤖 How to Set Up "Extract Text from Images/PDFs" - View Guide Here
Related Keywords
Related Keywords
extract data, AI error, data extraction, harmful content, text extraction, manage and transform data