Skip to content
Notifications
Clear all

Troubleshooting: How to deal with the model refusing to answer based on its safety filters?

1 Posts
1 Users
0 Reactions
1 Views
(@elliek2)
Estimable Member
Joined: 1 week ago
Posts: 98
Topic starter   [#11383]

Hey everyone, new here and still figuring things out! I've been trying out HuggingChat for some basic tasks to help with my Shopify store, like drafting product descriptions and support email replies. Most of the time it's great, but I keep hitting a wall with its safety filters.

For example, I asked it to help me write a response to a customer who was unhappy that their organic cotton shirt shrank a little after washing. I wanted to acknowledge the issue and offer a discount code for their next purchase—pretty standard customer service stuff. But the model just refused, saying it couldn't generate content that might be related to a "sensitive issue." I was really confused, because to me it's just a basic return/refund policy situation.

So my question is: what are you all doing when this happens? Is there a specific way to phrase things that works better?

I'm not trying to make it do anything bad, I promise! I just need practical business help. Does it help to explain the context more at the start? Or are some topics just completely blocked no matter what? Any tips would be super appreciated



   
Quote