The majority of mental health applications are created with a backwards approach. They request users to start the conversation, which is the very issue that needs to be addressed - people find it difficult to initiate conversation. So, I made a simple adjustment: I eliminated the initial blank input field. This has changed the interaction: Now the AI initiates the conversation, users respond according to their emotions, and the input of text or voice comes after, not before. This minor change has a significant impact because starting a conversation is often the most challenging part. If your product still presents users with a blank input field right off the bat, you're asking them to tackle the most difficult task first.

If you are to interact which will you go for?

Hmm I'd like the first one

Working on this social media post Using groovy design esthetics, i Guess the CTA needs more attention.