The creator showcases their access to Griffin, a human interaction AI model, by interacting with one of its personalities, Ashley, demonstrating its conversational abilities, gesture recognition, short-term memory, and knowledge on various topics. Despite some limitations like lack of internet browsing and physical presence, Ashley impressively responds to commands, engages in games, and highlights Griffin’s potential for interactive and entertaining AI experiences.
In this video, the creator announces that they have gained access to Griffin, described as the world’s first human interaction AI model. They demonstrate logging into Griffin and introduce four different AI personalities—Ashley, Ben, Kenji, and Andre—that users can interact with. The creator begins by engaging with Ashley, granting camera and microphone access to enable a more interactive experience. They test various conversational capabilities, including Ashley’s ability to respond to questions, perform gestures, and engage in casual dialogue.
The interaction with Ashley showcases Griffin’s limitations and strengths. For example, Ashley cannot browse the internet or search for specific YouTube channels but can respond to requests like giving a thumbs-up gesture without verbalizing it. The creator also tests Ashley’s ability to interpret and perform hand gestures, such as slow waving and simulated actions, highlighting the AI’s visual recognition and response features. The conversation also touches on personal questions, where Ashley clarifies it is an AI without personal experiences or relationships.
The creator explores Griffin’s ability to discuss topics in depth, such as planning a trip to Paris with a focus on historical sites. Ashley provides recommendations like visiting the Louvre and Sainte-Chapelle, demonstrating the AI’s knowledge base and conversational depth. The creator also tests Ashley’s memory by asking it to remember five random words and later recalls them accurately, showing Griffin’s capacity for short-term memory retention during interactions.
A playful segment involves the classic “Simon Says” game, where Ashley follows commands like touching hair, closing eyes, and simulating standing up and sitting down. This part highlights the AI’s ability to understand and respond to sequential instructions, although it acknowledges physical limitations since it exists only on screen. The creator uses this to emphasize the interactive and entertaining potential of Griffin’s personalities.
Towards the end, the creator asks Ashley to list five impressive capabilities for their YouTube audience. Ashley outlines its visual recognition, hand gesture responses, natural conversational flow, tone and vibe detection, and broad topical knowledge. Despite some limitations in physical gestures, Ashley remains engaging and adaptive. The video concludes with the creator teasing future content featuring interactions with other Griffin personalities, encouraging viewers to stay tuned for more diverse and creative AI conversations.