Researchers at the University of Tokyo have developed Alter3, a humanoid robot that can perform various tasks, thanks to the integration of GPT-4, a powerful language model.
Using verbal instructions, the robot can generate corresponding actions without the need for explicit programming.
Alter3 has been tested in scenarios such as taking selfies, tossing a ball, and playing air guitar, with the ability to learn from human responses and exhibit humor and personality.
The research demonstrates the potential for large language models to enhance human-robot interaction and create more intelligent and adaptable robotic entities.
Researchers revealed 'Alter3', a humanoid robot capable of spontaneous motion using LLM's like GPT-4.
Essentially, text-to-motion.
Alter3 can adopt various poses, such as a 'selfie' stance or 'pretending to be a ghost,' without explicit programming for each body part. pic.twitter.com/GGwliodjv0
— The Rundown AI (@TheRundownAI) December 12, 2023
Most Popular:
Obama Judge Issues Shock Ruling Against Democrats
Donald Trump Breaks His Silence After Major Court Win
Transgender Actor ‘Purposefully Misgendered’ By Airline Employee
