Ground Truth Ep.3 | Prof. Prathosh A.P, IISC Bangalore

Prof. Prathosh A. P. (Electrical Communication Engineering, IISc Bangalore) grew up in Mysore without knowing that IITs existed, finished his PhD at IISc by 25, taught deep learning at IIT Delhi, sold his first startup and now runs a second, LatentForce. After his classroom talk on AI and education went viral, he sat down with us for an honest conversation about what changes when intelligence stops being valuable: why first-principles thinking and storytelling matter more than ever, what we should teach our children, whether AI can be conscious, and why India has to own its AI models. Hosted by Vizuara co-founders Dr. Rajat Dandekar, Dr. Raj Dandekar and Dr. Sreedath Panat. We also cover how an academic sells a company, the "pet dog versus street dog" choice, his 15 years of nightly philosophy classes, how he hires for intent, and what research means when coding agents can write the code. CHAPTERS 0:00 Cold open 1:28 Introductions 3:14 From Mysore to a PhD at 25 9:19 Why every PhD should intern 11:54 Bringing deep learning to IIT Delhi 13:21 Math is the language of AI 19:08 Research that people use 23:58 First-principles thinking 26:52 An academic sells a company 30:08 Why PhDs struggle to sell 35:44 Does good work speak for itself? 37:32 When intelligence is commoditised 41:15 Finding mentors outside the IITs 48:30 Will we trust AI over people? 50:56 Storytelling and value 53:37 What should our children learn? 1:00:58 Universities in the age of agents 1:06:32 15 years of philosophy, every night 1:09:50 Can AI be conscious? 1:19:27 The AI divide 1:21:33 Should India build frontier models? 1:31:31 Money, talent and the DeepSeek gap 1:37:53 Building LatentForce 1:42:10 Hiring for intent 1:45:20 What should students learn now? 1:49:01 Research when agents write the code 1:59:12 Closing thoughts LINKS Prof. Prathosh A. P., IISc: https://ece.iisc.ac.in/ LatentForce: https://www.latentforce.ai/ Ground Truth Episode 2 with Prof. B. Ravindran: https://ground-truth.vizuara.ai Ground Truth by Vizuara: https://ground-truth.vizuara.ai PAPERS AND WORK MENTIONED (1) Hinton, Vinyals & Dean, Distilling the Knowledge in a Neural Network (2015): https://arxiv.org/abs/1503.02531 (2) DeepSeek-V2 (Multi-head Latent Attention): https://arxiv.org/abs/2405.04434 (3) Qwen Technical Report: https://arxiv.org/abs/2309.16609 (4) Backpropagation: https://en.wikipedia.org/wiki/Backpropagation (5) CogniAble: https://www.cogniable.tech/ (6) MIT delta v accelerator: https://entrepreneurship.mit.edu/accelerator/ PHOTO CREDITS (Wikimedia Commons) IIT Delhi main building: TheVyomanaut, CC BY-SA 4.0 · Walter Lewin: Qwsxza123890, CC BY-SA 4.0 · Yuval Noah Harari: Martin Kraft, CC BY-SA 4.0 · Gregor Mendel, Abraham Lincoln (Alexander Gardner), Mahatma Gandhi (Elliott & Fry), Socrates (Louvre): public domain. Logos are trademarks of their owners. Web pages and papers shown for reference. Ground Truth by Vizuara: honest conversations with the people shaping AI in India. Making sense of the noise.