make sure that gpt-4o is in your allowed models comma separate list in the dotAI app configuration screen
make sure you have created an embedded content index and have content in the index you are trying to use. To do this, you can go to the dotAI screen > Manage Embeddings / Indexes and add some content to the default index using a content query like +contentType:* which will add all your content to the embedded index.
Again, in the dotAI > Content Search and Chat tab, try running a few queries against the index you have created. Open your developer tools and look at the network tab and the api call payload for an example of what a working query should look like
If none of this works, I would tail the application logs and see if there are any sepecific error messages. If so, feel free to share them here to help us help you.