If a chatbot that can carry on an intelligent conversation about itself doesn't have a 'semblance of consciousness' then the word 'semblance' is meaningless.
Moltbook demonstrates that AI models simply do not engage in behavior analogous to human behavior. Compare Moltbook to Reddit and the difference should be obvious.