Abstract: Artificial intelligence is rapidly reshaping economies... [Keep your existing abstract text here]

Bio: Dr. Kalai Ramea is an accomplished scientist... [Keep your existing bio text here]

Abstract: Current machine "intelligence" is built on a silent assumption: intelligence equals text. Tokenization, orthography, captioning layers, and text-aligned embeddings dominate our architectures. What is missing are waveform-native reasoning, prosody, tone, rhythm, gesture, silence, context density, and communal signal dynamics. Billions are audio-literate but not text-dominant. Current systems force speech into transcription bottlenecks, collapse visual meaning into captions, and discard paralinguistic entropy. The result is structural exclusion encoded at the representation layer. This keynote introduces Textless Audio & Visual Machine Intelligence: signal-first architectures that learn directly from spectral manifolds, visual fields, and distributional community signals without orthographic reduction. We formalize risk quantification directly in signal space and embed uncertainty into "causal agentic World Models" capable of counterfactual reasoning and decision-making. We further define X-to-X causal communication: waveform-to-vision, vision-to-action, signal-to-signal, bypassing textual intermediaries. The next frontier is not larger language models. It is a collective intelligence that hears, sees, models causality, and communicates, without requiring text to exist.

Bio:Hamidou Tembine is co-founder of Timadie and Professor of Machine Intelligence at the School of Engineering, University of Québec (Canada). He received a master's degree in Applied Mathematics from École Polytechnique, Palaiseau (France), a master's degree in game theory and economics, and a Ph.D. in computer science from INRIA and the University of Avignon. He is founding director of the Learning and Game Theory Laboratory and one of the principal investigators of the Center on Stability, Instability, and Turbulence. He has also co-founded Grabal, WETE, and AI Mali, and founded Guinaga, SK1 Sogoloton, and WETE. He is the author of more than 300 publications and several books, including. Distributed Strategic Learning for Engineers (CRC Press), Game Theory and Learning in Wireless Networks (Elsevier),Mean-Field-Type Games I-II and for Engineers, Machine Intelligence in Africa in 20 Questions, and GPT Meets Game Theory. He is a senior member of IEEE, recipient of the IEEE ComSoc Outstanding Young Researcher Award, and winner of more than ten best paper awards, all in game theory. He has been recognized as a Next Einstein Fellow (2017) and Simons Senior Fellow (2020). His current research interests span learning, evolution, and games, with applications in agriculture, food, water, energy, communications, transportation, healthcare, textless audio-to-audio machine intelligence and people-centered cyber-physical systems security.