We know you’ve been waiting for this – and it’s finally here! The AI Chatbot can now receive and analyze attachments, making conversation with your clients more interactive and personal than ever
Awesome! Does your new feature incorporate safety or security filters? Bad actors could (and will) upload files containing malware or viruses. Please advise.
The team is making the AI Chatbot more powerful with every release.
I wanted to ask the Elfsight Team and Community: What client-side use cases do you see for this feature, and how would you train the AI Chatbot to respond or ask follow-up questions?
I know many of us want to add photos into the knowledge base so the chatbot can show clients our ownimages. But the client-side experience is especially interesting when you start thinking about how the AI should ask for information.
Example #1: Customer Support: A client tells the AI Chatbot they’re having an issue with a product they purchased. The chatbot can automatically ask the client to upload a photo so the support team can review the issue early and speed up troubleshooting.
Curious to hear what other use cases people are thinking about.
Of the supported file formats, only PDFs could potentially carry malware. Currently, we don’t scan files for viruses or malware.
I understand your concern, and we’ll consider adding safety filters for PDFs in the future. In the meantime, you’re welcome to upvote this idea on our Wishlist - Add safety filters for PDF files sent during the chat
Thanks, Max. Raising awareness about potential malware delivery via the new attachment feature is critical. This risk affects everyone, including Elfsight’s servers.
Elfsight servers can’t be affected by these types of files. The only risk is for the end recipient (widget owner) of the transcript email with attachments and the environment where they open the file.
So if you notice any suspicious file in your transcript, we recommend not opening it.
