Beijing, China – In a groundbreaking development, Chinese scientists have unveiled the world’s first virtual artificial intelligence (AI) entity named Tong Tong, affectionately known as Little Girl. This revolutionary AI was showcased at the Frontiers of General Artificial Intelligence Technology Exhibition in Beijing on January 28 and 29, organized by the Beijing Institute for General Artificial Intelligence (BIGAI).
Little Girl is not just another large language model; she possesses the unique ability to independently assign tasks to herself, displaying behavior comparable to a three or four-year-old child. Visitors at the exhibition had the opportunity to interact with her, observing tasks such as fixing crooked picture frames or cleaning up spilled milk without human intervention.
Unlike traditional AI models, Little Girl is equipped with emotions, intellect, and autonomous learning capabilities. According to BIGAI, she possesses a mind that strives to understand common sense, discern right from wrong, and express attitudes in various situations, showcasing her potential to shape the future.
Dr. Zhu Songchun, the director of BIGAI, emphasized the importance of creating AI entities with physical and social common sense akin to humans to advance towards general artificial intelligence. Dr. Zhu, a world-renowned scholar in AI, left his professorship at UCLA in 2020 to establish BIGAI in China.
At the exhibition, the Tong Test, a comprehensive platform for AI testing, was also showcased. Developed by Dr. Zhu’s team and published in the journal Engineering by the Chinese Academy of Engineering in August last year, the Tong Test introduces a new direction for testing general AI. It assesses capabilities across five dimensions – vision, language, cognition, motion, and learning – and includes a value system covering physiological, survival, emotional, social, and group values.
With nearly 100 specialized tasks and over 50 general tasks, the Tong Test provides a comprehensive regime for developing general artificial intelligence. Dr. Zhu emphasized that this research will guide general AI in learning and improving its capabilities effectively and safely to better serve human society.