Many bloggers have been sharing AI tutorials recently It's difficult to return to the domestic dapp ecosystem after getting used to using GPT But I haven't had a chance to go to Gemini yet this time The main reason is that when experiencing AI based dapps, it is found that compatibility with nodes can cause a sensory gap For example, there may be occasional crashes or node failures during use Besides the original device, it also seems to be related to the large computational load of current AI dapps So this time I paid attention to @ EPHYRA_AI's update on the core points. Compared to the previous vague concepts, this time I finally transitioned to the somatosensory aspect that I am more concerned about The project has optimized the core somatosensory aspects of perception, emotion and cognition, behavior and expression, and self modeling Perception: able to hear, speak, and see. For example, when you speak, it can understand and respond to you with voice and facial expressions, just like face-to-face communication Emotions and Cognition: Characters can experience emotional fluctuations, and AI will have likes and dislikes, feelings of goodwill or anger towards you, and these emotions will not change instantly Behavior and expression: Presenting behavior characteristics that are in line with the situation, AI no longer responds like customer service Q&A, but will respond with appropriate tone according to the situation, such as being gentle when comforting and relaxed when joking Self model: Through long-term interaction, AI will develop its own personality, understand you better as we get along, and actively evolve over time, growing like a real partner After reviewing the project roadmap, there should be plans for a campaign in the future. When the time comes, there will be some points to focus on and keep an eye on