title
Jeff Hawkins: The Thousand Brains Theory of Intelligence | Lex Fridman Podcast #208

description
Jeff Hawkins is a neuroscientist and cofounder of Numenta, a neuroscience research company. Please support this podcast by checking out our sponsors: - Codecademy: https://codecademy.com and use code LEX to get 15% off - BiOptimizers: http://www.magbreakthrough.com/lex to get 10% off - ExpressVPN: https://expressvpn.com/lexpod and use code LexPod to get 3 months free - Eight Sleep: https://www.eightsleep.com/lex and use code LEX to get special savings - Blinkist: https://blinkist.com/lex and use code LEX to get 25% off premium EPISODE LINKS: A Thousand Brain (book): https://amzn.to/3AmxJt7 Numenta's Twitter: https://twitter.com/Numenta Numenta's Website: https://numenta.com/ PODCAST INFO: Podcast website: https://lexfridman.com/podcast Apple Podcasts: https://apple.co/2lwqZIr Spotify: https://spoti.fi/2nEwCF8 RSS: https://lexfridman.com/feed/podcast/ Full episodes playlist: https://www.youtube.com/playlist?list=PLrAXtmErZgOdP_8GztsuKi9nrraNbKKp4 Clips playlist: https://www.youtube.com/playlist?list=PLrAXtmErZgOeciFP3CBCIEElOJeitOr41 OUTLINE: 0:00 - Introduction 3:04 - Collective intelligence 9:46 - The origin of intelligence in the human brain 22:59 - How intelligent life evolved on Earth 33:58 - Why humans are special in the universe 37:16 - Neurons 41:30 - A Thousand Brains theory of intelligence 50:10 - How to build superintelligent AI 1:08:10 - Sam Harris and existential risk of AI 1:20:12 - Neuralink 1:27:02 - Will AI prevent the self-destruction of human civilization? 1:32:34 - Communicating human knowledge to alien civilizations 1:42:50 - Devil's advocate 1:47:45 - Human nature 1:56:07 - Hardware for AI 2:02:46 - Advice for young people SOCIAL: - Twitter: https://twitter.com/lexfridman - LinkedIn: https://www.linkedin.com/in/lexfridman - Facebook: https://www.facebook.com/lexfridman - Instagram: https://www.instagram.com/lexfridman - Medium: https://medium.com/@lexfridman - Reddit: https://reddit.com/r/lexfridman - Support on Patreon: https://www.patreon.com/lexfridman

detail
{'title': 'Jeff Hawkins: The Thousand Brains Theory of Intelligence | Lex Fridman Podcast #208', 'heatmap': [{'end': 7647.679, 'start': 7559.054, 'weight': 1}], 'summary': "Neuroscientist jeff hawkins introduces the 'a thousand brains' theory, discussing the preservation of human knowledge and the neocortex's role in high-level functions, including predictive perception and societal impact of ai, and explores the potential integration of ai and human consciousness.", 'chapters': [{'end': 46.491, 'segs': [{'end': 46.491, 'src': 'embed', 'start': 0.189, 'weight': 0, 'content': [{'end': 6.603, 'text': 'The following is a conversation with Jeff Hawkins, a neuroscientist seeking to understand the structure,', 'start': 0.189, 'duration': 6.414}, {'end': 9.369, 'text': 'function and origin of intelligence in the human brain.', 'start': 6.603, 'duration': 2.766}, {'end': 18.738, 'text': 'He previously wrote the seminal book on the subject titled On Intelligence, and recently a new book called A Thousand Brains,', 'start': 10.336, 'duration': 8.402}, {'end': 23.899, 'text': 'which presents a new theory of intelligence that Richard Dawkins, for example,', 'start': 18.738, 'duration': 5.161}, {'end': 28.48, 'text': 'has been raving about calling the book quote brilliant and exhilarating.', 'start': 23.899, 'duration': 4.581}, {'end': 34.141, 'text': "I can't read those two words and not think of him saying it in his British accent.", 'start': 29.34, 'duration': 4.801}, {'end': 41.726, 'text': 'Quick mention of our sponsors, Codecademy, BioOptimizers, ExpressVPN, Asleep, and Blinkist.', 'start': 34.98, 'duration': 6.746}, {'end': 44.689, 'text': 'Check them out in the description to support this podcast.', 'start': 42.127, 'duration': 2.562}, {'end': 46.491, 'text': 'As a side note,', 'start': 45.39, 'duration': 1.101}], 'summary': 'Neuroscientist jeff hawkins discusses new theory of intelligence in his book a thousand brains, endorsed by richard dawkins.', 'duration': 46.302, 'max_score': 0.189, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/Z1KwkpTUbkg/pics/Z1KwkpTUbkg189.jpg'}], 'start': 0.189, 'title': "Jeff hawkins' a thousand brains theory", 'summary': "Delves into a conversation with neuroscientist jeff hawkins, who introduces 'a thousand brains' theory, acclaimed by richard dawkins, and highlights sponsors including codecademy, biooptimizers, expressvpn, asleep, and blinkist.", 'chapters': [{'end': 46.491, 'start': 0.189, 'title': 'Jeff hawkins: a thousand brains theory', 'summary': "Features a conversation with neuroscientist jeff hawkins, author of 'a thousand brains', presenting a new theory of intelligence, lauded by richard dawkins, and mentions sponsors codecademy, biooptimizers, expressvpn, asleep, and blinkist.", 'duration': 46.302, 'highlights': ["Jeff Hawkins presents the new theory of intelligence in 'A Thousand Brains' lauded by Richard Dawkins.", 'Mentions sponsors: Codecademy, BioOptimizers, ExpressVPN, Asleep, and Blinkist.']}], 'duration': 46.302, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/Z1KwkpTUbkg/pics/Z1KwkpTUbkg189.jpg', 'highlights': ["Jeff Hawkins introduces 'A Thousand Brains' theory acclaimed by Richard Dawkins.", 'Mentions sponsors: Codecademy, BioOptimizers, ExpressVPN, Asleep, and Blinkist.']}, {'end': 1196.28, 'segs': [{'end': 76.092, 'src': 'embed', 'start': 46.491, 'weight': 5, 'content': [{'end': 54.798, 'text': 'let me say that one small but powerful idea that Jeff Hawkins mentions in his new book is that if human civilization were to destroy itself,', 'start': 46.491, 'duration': 8.307}, {'end': 58.201, 'text': 'all of knowledge, all our creations will go with us.', 'start': 54.798, 'duration': 3.403}, {'end': 67.246, 'text': "He proposes that we should think about how to save that knowledge in a way that long outlives us, whether that's on Earth, in orbit,", 'start': 59.141, 'duration': 8.105}, {'end': 76.092, 'text': 'around Earth or in deep space, and then to send messages that advertise this backup of human knowledge to other intelligent alien civilizations.', 'start': 67.246, 'duration': 8.846}], 'summary': 'Jeff hawkins proposes saving human knowledge to outlive us and advertising it to other intelligent alien civilizations.', 'duration': 29.601, 'max_score': 46.491, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/Z1KwkpTUbkg/pics/Z1KwkpTUbkg46491.jpg'}, {'end': 182.894, 'src': 'embed', 'start': 158.974, 'weight': 6, 'content': [{'end': 166.017, 'text': 'If we can record these experiences in the highest resolution directly from the human brain, such that aliens would be able to replay them.', 'start': 158.974, 'duration': 7.043}, {'end': 168.959, 'text': 'that is what we should store and send as a message.', 'start': 166.017, 'duration': 2.942}, {'end': 177.047, 'text': 'not Wikipedia, but the extremes of conscious experiences, the most important of which, of course, is love.', 'start': 169.739, 'duration': 7.308}, {'end': 182.894, 'text': 'This is the Lex Friedman Podcast, and here is my conversation with Jeff Hawkins.', 'start': 178.028, 'duration': 4.866}], 'summary': 'Record highest resolution human brain experiences, including love, for alien message.', 'duration': 23.92, 'max_score': 158.974, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/Z1KwkpTUbkg/pics/Z1KwkpTUbkg158974.jpg'}, {'end': 248.339, 'src': 'embed', 'start': 223.012, 'weight': 0, 'content': [{'end': 229.154, 'text': 'But we know from, one thing to know about brains is that when you think of things, you often erase the memory and rewrite it again.', 'start': 223.012, 'duration': 6.142}, {'end': 233.815, 'text': "So yes, but I have a memory of you and I have, that's instantiated in synapses.", 'start': 229.194, 'duration': 4.621}, {'end': 236.216, 'text': "There's a simpler way to think about it.", 'start': 234.235, 'duration': 1.981}, {'end': 241.977, 'text': 'We have a model of the world in your head, and that model is continually being updated.', 'start': 236.996, 'duration': 4.981}, {'end': 243.758, 'text': 'I updated this morning.', 'start': 242.657, 'duration': 1.101}, {'end': 246.698, 'text': 'You offered me this water that was from the refrigerator.', 'start': 243.818, 'duration': 2.88}, {'end': 248.339, 'text': 'I remember these things.', 'start': 247.459, 'duration': 0.88}], 'summary': 'Brains continuously update a model of the world through memory and rewriting.', 'duration': 25.327, 'max_score': 223.012, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/Z1KwkpTUbkg/pics/Z1KwkpTUbkg223012.jpg'}, {'end': 497.896, 'src': 'embed', 'start': 467.894, 'weight': 1, 'content': [{'end': 469.876, 'text': 'Then you could say well, how do brains interact with each other?', 'start': 467.894, 'duration': 1.982}, {'end': 471.757, 'text': 'And what is the nature of language?', 'start': 470.396, 'duration': 1.361}, {'end': 476.541, 'text': "And how do we share models that I've learned something about the world, how do I share it with you?", 'start': 472.458, 'duration': 4.083}, {'end': 480.063, 'text': 'Which is really what sort of communal intelligence is?', 'start': 476.561, 'duration': 3.502}, {'end': 481.544, 'text': 'I know something, you know something.', 'start': 480.243, 'duration': 1.301}, {'end': 483.906, 'text': "We've had different experiences in the world.", 'start': 482.345, 'duration': 1.561}, {'end': 486.708, 'text': "I've learned something about brains, maybe I can impart that to you.", 'start': 484.507, 'duration': 2.201}, {'end': 490.071, 'text': "You've learned something about physics and you can impart that to me.", 'start': 486.748, 'duration': 3.323}, {'end': 497.896, 'text': 'but it all comes down to even just the epistemological question of, well, what is knowledge and how do you represent it?', 'start': 491.832, 'duration': 6.064}], 'summary': 'Exploring how brains interact, share knowledge, and represent it.', 'duration': 30.002, 'max_score': 467.894, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/Z1KwkpTUbkg/pics/Z1KwkpTUbkg467894.jpg'}, {'end': 547.022, 'src': 'embed', 'start': 521.653, 'weight': 2, 'content': [{'end': 530.576, 'text': "I mean we can't deny that the brain is the core element here in at least I think it's obvious the brain is the core element in all theories of intelligence.", 'start': 521.653, 'duration': 8.923}, {'end': 532.737, 'text': "It's where knowledge is represented.", 'start': 531.337, 'duration': 1.4}, {'end': 534.278, 'text': "It's where knowledge is created.", 'start': 532.817, 'duration': 1.461}, {'end': 541.261, 'text': "We interact, we share, we build upon each other's work, but without a brain, you'd have nothing.", 'start': 535.298, 'duration': 5.963}, {'end': 546.182, 'text': 'There would be no intelligence without brains.', 'start': 542.041, 'duration': 4.141}, {'end': 547.022, 'text': "That's where we start.", 'start': 546.242, 'duration': 0.78}], 'summary': 'The brain is the core of intelligence theories, where knowledge is created and represented.', 'duration': 25.369, 'max_score': 521.653, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/Z1KwkpTUbkg/pics/Z1KwkpTUbkg521653.jpg'}, {'end': 676.672, 'src': 'embed', 'start': 639.273, 'weight': 3, 'content': [{'end': 644.935, 'text': "And so each, what we call the column in the cortex, there's about 150,000 of them, is a complete modeling system.", 'start': 639.273, 'duration': 5.662}, {'end': 648.779, 'text': "So it's a collective intelligence in your head in some sense.", 'start': 645.675, 'duration': 3.104}, {'end': 652.743, 'text': 'So the thousand brains theory says well, where do I have knowledge about?', 'start': 649.359, 'duration': 3.384}, {'end': 655.506, 'text': 'you know this coffee cup or where is the model of this cell phone?', 'start': 652.743, 'duration': 2.763}, {'end': 656.848, 'text': "It's not in one place.", 'start': 655.987, 'duration': 0.861}, {'end': 661.113, 'text': "It's in thousands of separate models that are complementary and they communicate with each other through voting.", 'start': 656.988, 'duration': 4.125}, {'end': 666.163, 'text': "So this idea that we have, we feel like we're one person That's our experience.", 'start': 661.613, 'duration': 4.55}, {'end': 666.984, 'text': 'We can explain that.', 'start': 666.243, 'duration': 0.741}, {'end': 671.167, 'text': "But reality, there's lots of these, it's almost like little brains.", 'start': 667.284, 'duration': 3.883}, {'end': 676.672, 'text': "They're sophisticated modeling systems, about 150,000 of them in each human brain.", 'start': 671.187, 'duration': 5.485}], 'summary': 'Each cortex column contains about 150,000 modeling systems forming a collective intelligence in the brain, demonstrating the thousand brains theory.', 'duration': 37.399, 'max_score': 639.273, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/Z1KwkpTUbkg/pics/Z1KwkpTUbkg639273.jpg'}], 'start': 46.491, 'title': 'Preserving human knowledge and collective intelligence', 'summary': "Discusses preserving human knowledge for potential alien civilizations, while also exploring the brain's ability to form models of the world and introducing the thousand brains theory of intelligence, which suggests the neocortex consists of thousands of independent modeling systems.", 'chapters': [{'end': 182.894, 'start': 46.491, 'title': 'Preserving human knowledge for aliens', 'summary': 'Discusses the idea of preserving human knowledge for long-term survival, proposing the storage and transmission of essential human experiences and consciousness to potential alien civilizations, rather than just our technological and scientific achievements.', 'duration': 136.403, 'highlights': ['The proposal to save human knowledge in a way that outlives us and to transmit messages to potential alien civilizations is explored. Jeff Hawkins suggests the need to consider preserving human knowledge in a way that outlives us, whether on Earth, in orbit, around Earth, or in deep space, and to send messages advertising this backup to potential alien civilizations.', 'The thought experiment prompts the consideration of what aspects of human knowledge and experiences are essential and likely to outlive us. The chapter explores the deeper question of what information in human knowledge is essential and likely to outlive us, prompting consideration of achievements and experiences such as complex buildings, science, music, art, consciousness, and subjective experiences.', 'The concept of preserving and transmitting the extremes of conscious experiences, particularly the emotion of love, is emphasized as a crucial aspect of human knowledge to be stored and sent as a message to potential alien civilizations. The chapter emphasizes the importance of preserving and transmitting the extremes of conscious experiences, focusing on the emotion of love as a crucial aspect of human knowledge to be stored and transmitted to potential alien civilizations, rather than just technological or scientific achievements.']}, {'end': 584.473, 'start': 184.266, 'title': 'Understanding the brain and collective intelligence', 'summary': "Explores the brain's ability to form and update models of the world, including memories of individuals, and emphasizes the fundamental role of the neocortex in understanding and interacting with the world, while also touching upon the concept of collective intelligence and the brain's role in knowledge representation.", 'duration': 400.207, 'highlights': ['The neocortex, comprising 75% of the brain, plays a fundamental role in understanding and interacting with the world, as it continuously updates a model that includes memories of individuals and objects. The neocortex, comprising 75% of the brain, plays a fundamental role in understanding and interacting with the world, as it continuously updates a model that includes memories of individuals and objects.', "The brain's ability to create, update, and share models of the world underlies the concept of collective intelligence, with language being a key part of this capability. The brain's ability to create, update, and share models of the world underlies the concept of collective intelligence, with language being a key part of this capability.", 'The brain is deemed as the essential component in theories of intelligence, serving as the core element where knowledge is represented and created, highlighting the significance of neurons and synapses in understanding fundamental questions about knowledge and cognition. The brain is deemed as the essential component in theories of intelligence, serving as the core element where knowledge is represented and created, highlighting the significance of neurons and synapses in understanding fundamental questions about knowledge and cognition.']}, {'end': 1196.28, 'start': 587.064, 'title': 'Thousand brains theory: collective intelligence in the neocortex', 'summary': 'Introduces the thousand brains theory of intelligence, which proposes that the neocortex is comprised of thousands of independent modeling systems that communicate through voting, allowing the brain to collectively perceive and learn models of the world, with each column in the cortex containing about 150,000 complete modeling systems.', 'duration': 609.216, 'highlights': ['The neocortex is comprised of thousands of independent modeling systems that communicate through voting, allowing the brain to collectively perceive and learn models of the world. neocortex, independent modeling systems, voting, collective perception, learning models', "Each column in the cortex contains about 150,000 complete modeling systems, presenting a total different way of thinking about the brain's structure. cortex, complete modeling systems, 150,000 systems per column, brain's structure", 'The essence of intelligence is the ability to learn a model of the world, representing the structure and behavior of objects, acquired through interaction and movement. intelligence, learning model, representation of structure and behavior, interaction and movement', 'Prediction is fundamental to intelligence as it allows the model to correct itself by learning where it is wrong, enabling the brain to update and refine its internal representations. prediction, model correction, updating internal representations']}], 'duration': 1149.789, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/Z1KwkpTUbkg/pics/Z1KwkpTUbkg46491.jpg', 'highlights': ['The neocortex, comprising 75% of the brain, continuously updates a model that includes memories of individuals and objects.', "The brain's ability to create, update, and share models of the world underlies the concept of collective intelligence, with language being a key part of this capability.", 'The brain is deemed as the essential component in theories of intelligence, serving as the core element where knowledge is represented and created, highlighting the significance of neurons and synapses in understanding fundamental questions about knowledge and cognition.', 'The neocortex consists of thousands of independent modeling systems that communicate through voting, allowing the brain to collectively perceive and learn models of the world.', "Each column in the cortex contains about 150,000 complete modeling systems, presenting a total different way of thinking about the brain's structure.", 'The proposal to save human knowledge in a way that outlives us and to transmit messages to potential alien civilizations is explored.', 'The concept of preserving and transmitting the extremes of conscious experiences, particularly the emotion of love, is emphasized as a crucial aspect of human knowledge to be stored and sent as a message to potential alien civilizations.', 'The neocortex is deemed as the essential component in theories of intelligence, serving as the core element where knowledge is represented and created.']}, {'end': 2213.616, 'segs': [{'end': 1243.799, 'src': 'embed', 'start': 1196.32, 'weight': 1, 'content': [{'end': 1199.422, 'text': "Just by looking at it and playing around with it, I'd update it and say, this is a new type of water bottle.", 'start': 1196.32, 'duration': 3.102}, {'end': 1209.307, 'text': "So you're talking about sort of complicated things like a water bottle, but this also applies for just basic vision, just like seeing things.", 'start': 1200.12, 'duration': 9.187}, {'end': 1214.67, 'text': "It's almost like a precondition of just perceiving the world as predicting it.", 'start': 1210.107, 'duration': 4.563}, {'end': 1221.035, 'text': "It's just everything that you see is first passed through your prediction.", 'start': 1215.451, 'duration': 5.584}, {'end': 1223.076, 'text': 'Everything you see and feel.', 'start': 1221.115, 'duration': 1.961}, {'end': 1226.278, 'text': 'In fact, this was the insight I had back in the late 80s,', 'start': 1223.096, 'duration': 3.182}, {'end': 1234.935, 'text': 'and other people have reached the same idea is that every sensory input you get not just vision,', 'start': 1229.213, 'duration': 5.722}, {'end': 1241.158, 'text': 'but touch and hearing you have an expectation about it and a prediction.', 'start': 1234.935, 'duration': 6.223}, {'end': 1243.799, 'text': "Sometimes you can predict very accurately, sometimes you can't.", 'start': 1241.518, 'duration': 2.281}], 'summary': 'Sensory perception involves prediction, as seen in water bottle example, with an expectation and prediction for every sensory input.', 'duration': 47.479, 'max_score': 1196.32, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/Z1KwkpTUbkg/pics/Z1KwkpTUbkg1196320.jpg'}, {'end': 1310.13, 'src': 'embed', 'start': 1284.439, 'weight': 2, 'content': [{'end': 1290.502, 'text': 'i it really honestly feels that prediction is fundamental to everything.', 'start': 1284.439, 'duration': 6.063}, {'end': 1295.083, 'text': 'uh, to the way our mind operates, to intelligence, so like.', 'start': 1290.502, 'duration': 4.581}, {'end': 1296.664, 'text': "it's just a different way to see.", 'start': 1295.083, 'duration': 1.581}, {'end': 1301.406, 'text': 'intelligence, which is like everything, starts at prediction, and prediction requires a model.', 'start': 1296.664, 'duration': 4.742}, {'end': 1304.865, 'text': "You can't predict something unless you have a model of it.", 'start': 1302.162, 'duration': 2.703}, {'end': 1307.528, 'text': 'Right, but the action is prediction.', 'start': 1305.185, 'duration': 2.343}, {'end': 1310.13, 'text': 'The thing the model does is prediction.', 'start': 1308.469, 'duration': 1.661}], 'summary': 'Prediction is fundamental to intelligence, starting with a model for accurate prediction.', 'duration': 25.691, 'max_score': 1284.439, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/Z1KwkpTUbkg/pics/Z1KwkpTUbkg1284439.jpg'}, {'end': 1500.5, 'src': 'embed', 'start': 1473.499, 'weight': 4, 'content': [{'end': 1480.166, 'text': 'So early on there was a pressure to start understanding your environment, like where am I and where have I been?', 'start': 1473.499, 'duration': 6.667}, {'end': 1482.149, 'text': 'And what happened in those different places?', 'start': 1480.467, 'duration': 1.682}, {'end': 1487.175, 'text': 'So we still have this neural mechanism in our brains.', 'start': 1483.513, 'duration': 3.662}, {'end': 1493.097, 'text': "In the mammals, it's in the hippocampus and entorhinal cortex.", 'start': 1488.015, 'duration': 5.082}, {'end': 1494.338, 'text': 'These are older parts of the brain.', 'start': 1493.117, 'duration': 1.221}, {'end': 1497.119, 'text': 'And these are very well studied.', 'start': 1495.698, 'duration': 1.421}, {'end': 1500.5, 'text': 'We build a map of our environment.', 'start': 1498.359, 'duration': 2.141}], 'summary': 'Neural mechanism in brains builds maps of environments, studied in hippocampus and entorhinal cortex.', 'duration': 27.001, 'max_score': 1473.499, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/Z1KwkpTUbkg/pics/Z1KwkpTUbkg1473499.jpg'}, {'end': 1922.349, 'src': 'embed', 'start': 1898.749, 'weight': 0, 'content': [{'end': 1906.155, 'text': 'Yeah, so what is in your cortex and where does it fit in the various aspect of what the brain does? Like, how important is it to you??', 'start': 1898.749, 'duration': 7.406}, {'end': 1914.402, 'text': "Well, obviously again I mentioned again in the beginning it's about 70 to 75% of the volume of a human brain.", 'start': 1906.676, 'duration': 7.726}, {'end': 1917.165, 'text': 'So it dominates our brain in terms of size.', 'start': 1914.722, 'duration': 2.443}, {'end': 1919.907, 'text': 'Not in terms of number of neurons, but in terms of size.', 'start': 1917.665, 'duration': 2.242}, {'end': 1922.349, 'text': "Size isn't everything, Jeff.", 'start': 1920.788, 'duration': 1.561}], 'summary': "The cortex comprises 70-75% of the human brain's volume and dominates in size.", 'duration': 23.6, 'max_score': 1898.749, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/Z1KwkpTUbkg/pics/Z1KwkpTUbkg1898749.jpg'}], 'start': 1196.32, 'title': 'Predictive perception and intelligence', 'summary': "Explores the concept of predictive perception, emphasizing sensory inputs preconceived and tested against expectations, originated in the late 80s. it discusses the fundamental role of prediction in intelligence, the necessity of a model for prediction, and the neocortex's role in high-level functions, accounting for 70-75% of the human brain's volume.", 'chapters': [{'end': 1284.439, 'start': 1196.32, 'title': 'Predictive nature of perception', 'summary': 'Explores the concept of predictive perception, emphasizing how sensory inputs are preconceived and tested against expectations, with the insight originating in the late 80s, illustrating the constant background predictions for all senses.', 'duration': 88.119, 'highlights': ['The insight of sensory inputs being preconceived and tested against expectations originated in the late 80s, demonstrating constant background predictions for all senses.', 'Perception is compared to a test of understanding, where predictions serve as a measure of comprehension and learning.', 'The concept of predictive perception is exemplified through the analogy of identifying a water bottle by testing expectations against actual sensory input.']}, {'end': 1813.856, 'start': 1284.439, 'title': 'The role of prediction in intelligence', 'summary': "Discusses the fundamental role of prediction in intelligence, starting with the necessity of a model for prediction and expanding to the evolutionary advantage of understanding one's environment, leading to the theory of the origin of intelligence and the presence of equivalent modeling systems in each column of the neocortex.", 'duration': 529.417, 'highlights': ['The necessity of a model for prediction is highlighted as the starting point for intelligence, emphasizing that all actions are, in essence, predictions based on models.', "The evolutionary advantage of understanding one's environment is discussed, focusing on the pressure for early living things to intelligently move and the neural mechanisms in mammals for building a map of the environment.", "The theory of the origin of intelligence is presented with the proposal that the human brain's flexibility and ability to learn without prior knowledge suggests the existence of a universal learning algorithm, which likely originated from something more specific.", 'The presence of equivalent modeling systems in each column of the neocortex is highlighted as a major predictive aspect of the theory, indicating the existence of grid cell place cell equivalent cells working on the same principles as modeling systems in the brain.']}, {'end': 2213.616, 'start': 1813.856, 'title': 'Understanding the neocortex and human intelligence', 'summary': "Discusses the neocortex, which accounts for 70-75% of the human brain's volume, and its role in high-level functions such as vision, language, planning, and thinking, while also delving into the complexity of understanding collective intelligence and the uniqueness of human knowledge.", 'duration': 399.76, 'highlights': ['The neocortex accounts for 70-75% of the volume of a human brain, dominating in size and playing a crucial role in high-level functions like vision, language, planning, and thinking.', 'The Santa Fe Institute was founded to study complexity theory, reflecting the challenges in understanding complex systems and the minimal progress made in this field.', 'Human brains possess unique knowledge about the universe, DNA, and the origins of species, setting them apart from other species on Earth.', 'Understanding the neocortex is essential in comprehending human intelligence, although interactions with other parts of the brain influence how the neocortex manifests in human behavior and decision-making.', 'The chapter also touches on the challenges in understanding communal intelligence and the notion that humans may not hold a significant place on the complexity spectrum.']}], 'duration': 1017.296, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/Z1KwkpTUbkg/pics/Z1KwkpTUbkg1196320.jpg', 'highlights': ['The neocortex accounts for 70-75% of the volume of a human brain, dominating in size and playing a crucial role in high-level functions like vision, language, planning, and thinking.', 'The insight of sensory inputs being preconceived and tested against expectations originated in the late 80s, demonstrating constant background predictions for all senses.', 'The necessity of a model for prediction is highlighted as the starting point for intelligence, emphasizing that all actions are, in essence, predictions based on models.', 'The concept of predictive perception is exemplified through the analogy of identifying a water bottle by testing expectations against actual sensory input.', "The evolutionary advantage of understanding one's environment is discussed, focusing on the pressure for early living things to intelligently move and the neural mechanisms in mammals for building a map of the environment."]}, {'end': 3322.912, 'segs': [{'end': 2431.813, 'src': 'embed', 'start': 2389.496, 'weight': 0, 'content': [{'end': 2392.177, 'text': "So that's where we think that most of these predictions are internal.", 'start': 2389.496, 'duration': 2.681}, {'end': 2395.538, 'text': "That's why you can't be, they're internal neuron, you can't perceive them.", 'start': 2392.197, 'duration': 3.341}, {'end': 2400.839, 'text': 'Well, from understanding the prediction mechanism of a single neuron,', 'start': 2396.278, 'duration': 4.561}, {'end': 2407.82, 'text': "do you think there's deep insights to be gained about the prediction capabilities of the mini brains and then the bigger brain and the brain?", 'start': 2400.839, 'duration': 6.981}, {'end': 2408.52, 'text': 'Oh yeah, yeah, yeah.', 'start': 2407.84, 'duration': 0.68}, {'end': 2411.661, 'text': 'So having a prediction inside of their individual neuron is not that useful.', 'start': 2408.8, 'duration': 2.861}, {'end': 2413.273, 'text': 'So what??', 'start': 2412.932, 'duration': 0.341}, {'end': 2423.479, 'text': 'The way it manifests itself in neural tissue is that when a neuron emits these spikes a very singular type event.', 'start': 2415.174, 'duration': 8.305}, {'end': 2429.652, 'text': "If a neuron is predicting that it's going to be active, it emits its spike very a little bit sooner,", 'start': 2424.289, 'duration': 5.363}, {'end': 2431.813, 'text': 'just a few milliseconds sooner than it would have otherwise.', 'start': 2429.652, 'duration': 2.161}], 'summary': 'Studying neuron prediction mechanisms may enhance understanding of brain prediction capabilities.', 'duration': 42.317, 'max_score': 2389.496, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/Z1KwkpTUbkg/pics/Z1KwkpTUbkg2389496.jpg'}, {'end': 2519.173, 'src': 'embed', 'start': 2487.781, 'weight': 3, 'content': [{'end': 2490.703, 'text': 'So the theory of the thousand brains.', 'start': 2487.781, 'duration': 2.922}, {'end': 2493.964, 'text': 'if you were to count the number of brains, how would you do it?', 'start': 2490.703, 'duration': 3.261}, {'end': 2506.289, 'text': 'The thousand brain theory says that basically every cortical column in your neocortex is a complete modeling system and that when I ask where do I have a model of something like a coffee cup,', 'start': 2494.906, 'duration': 11.383}, {'end': 2508.71, 'text': "it's not in one of those models, it's in thousands of those models.", 'start': 2506.289, 'duration': 2.421}, {'end': 2510.73, 'text': "There's thousands of models of coffee cups.", 'start': 2509.09, 'duration': 1.64}, {'end': 2512.511, 'text': "That's what the thousand brains theory says.", 'start': 2511.03, 'duration': 1.481}, {'end': 2513.671, 'text': "Then there's a voting mechanism.", 'start': 2512.551, 'duration': 1.12}, {'end': 2519.173, 'text': "Then there's a voting mechanism, which is the thing which you're conscious of, which leads to your singular perception.", 'start': 2513.751, 'duration': 5.422}], 'summary': 'The thousand brain theory proposes that each cortical column in the neocortex is a complete modeling system, with thousands of models of objects like a coffee cup, leading to a singular perception through a voting mechanism.', 'duration': 31.392, 'max_score': 2487.781, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/Z1KwkpTUbkg/pics/Z1KwkpTUbkg2487781.jpg'}, {'end': 2669.143, 'src': 'embed', 'start': 2640.937, 'weight': 4, 'content': [{'end': 2643.163, 'text': 'And to do that, I have to have a reference frame for the coffee cup.', 'start': 2640.937, 'duration': 2.226}, {'end': 2645.409, 'text': 'It has to have a way of representing the location of my finger.', 'start': 2643.223, 'duration': 2.186}, {'end': 2647.557, 'text': 'the coffee cup.', 'start': 2646.537, 'duration': 1.02}, {'end': 2651.498, 'text': 'and then we realized of course, every part of your skin has to have a reference frame relative to things that touch.', 'start': 2647.557, 'duration': 3.941}, {'end': 2652.999, 'text': 'and then we did the same thing with vision.', 'start': 2651.498, 'duration': 1.501}, {'end': 2657.2, 'text': 'but so the idea that a reference frame is necessary to make a prediction.', 'start': 2652.999, 'duration': 4.201}, {'end': 2662.081, 'text': "when you're touching something or when you're seeing something and you're moving your eyes, removing your fingers,", 'start': 2657.2, 'duration': 4.881}, {'end': 2664.962, 'text': "it's just a requirement to know what to predict.", 'start': 2662.081, 'duration': 2.881}, {'end': 2668.182, 'text': "if i have a, if i have a structure, i'm going to make a prediction, i have to.", 'start': 2664.962, 'duration': 3.22}, {'end': 2669.143, 'text': 'i have to know where it is.', 'start': 2668.182, 'duration': 0.961}], 'summary': 'Reference frames are crucial for making predictions when interacting with objects.', 'duration': 28.206, 'max_score': 2640.937, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/Z1KwkpTUbkg/pics/Z1KwkpTUbkg2640937.jpg'}, {'end': 2960.06, 'src': 'embed', 'start': 2923.357, 'weight': 5, 'content': [{'end': 2931.084, 'text': 'So anyway, the modeling system that exists in every cortical column learns the hierarchical structure of objects.', 'start': 2923.357, 'duration': 7.727}, {'end': 2934.406, 'text': "So it's a very sophisticated modeling system in this grain of rice.", 'start': 2931.744, 'duration': 2.662}, {'end': 2937.567, 'text': "It's hard to imagine, but this grain of rice can do really sophisticated things.", 'start': 2934.466, 'duration': 3.101}, {'end': 2938.928, 'text': "It's got 100,000 neurons in it.", 'start': 2937.607, 'duration': 1.321}, {'end': 2941.509, 'text': "It's very sophisticated.", 'start': 2940.389, 'duration': 1.12}, {'end': 2949.394, 'text': 'So that same mechanism that can model a water bottle or a coffee cup can model conceptual objects as well.', 'start': 2941.93, 'duration': 7.464}, {'end': 2955.417, 'text': "That's the beauty of this discovery that this guy, Vernon Mountcastle, made many, many years ago,", 'start': 2949.654, 'duration': 5.763}, {'end': 2960.06, 'text': "which is that there's a single cortical algorithm underlying everything we're doing.", 'start': 2955.417, 'duration': 4.643}], 'summary': 'Cortical column model can learn and model objects, with 100,000 neurons, underlying various concepts.', 'duration': 36.703, 'max_score': 2923.357, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/Z1KwkpTUbkg/pics/Z1KwkpTUbkg2923357.jpg'}, {'end': 3048.347, 'src': 'embed', 'start': 3022.03, 'weight': 8, 'content': [{'end': 3032.36, 'text': 'So, looking at the brain, Is this a marvel of engineering, or is it pretty dumb stuff stacked on top of each other over a pretty extensive copy?', 'start': 3022.03, 'duration': 10.33}, {'end': 3033.12, 'text': "Can't it be both??", 'start': 3032.4, 'duration': 0.72}, {'end': 3034.761, 'text': "Can't it be both right??", 'start': 3033.26, 'duration': 1.501}, {'end': 3043.265, 'text': "I don't know if it can be both, because if it's an incredible engineering job, that means it's.", 'start': 3035.761, 'duration': 7.504}, {'end': 3044.865, 'text': 'So, evolution did a lot of work.', 'start': 3043.265, 'duration': 1.6}, {'end': 3048.347, 'text': 'Yeah, but then it just copied that.', 'start': 3045.566, 'duration': 2.781}], 'summary': 'Debating whether the brain is both a marvel of engineering and dumb stuff stacked on top of each other, with a focus on the evolutionary aspect.', 'duration': 26.317, 'max_score': 3022.03, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/Z1KwkpTUbkg/pics/Z1KwkpTUbkg3022030.jpg'}, {'end': 3128.173, 'src': 'embed', 'start': 3096.714, 'weight': 9, 'content': [{'end': 3100.415, 'text': "There's still some engineering problems to solve, but we know enough.", 'start': 3096.714, 'duration': 3.701}, {'end': 3105.516, 'text': 'And this is not like, oh, this is an interesting idea, we have to go think about it for another few decades.', 'start': 3101.315, 'duration': 4.201}, {'end': 3107.577, 'text': 'No, we actually understand it in pretty well details.', 'start': 3105.596, 'duration': 1.981}, {'end': 3111.118, 'text': 'So not all the details, but most of them.', 'start': 3108.317, 'duration': 2.801}, {'end': 3114.87, 'text': "So it's complicated, but it is an engineering problem.", 'start': 3111.949, 'duration': 2.921}, {'end': 3117.37, 'text': 'So in my company, we are working on that.', 'start': 3115.49, 'duration': 1.88}, {'end': 3120.171, 'text': 'We are basically laying out a roadmap, how we do this.', 'start': 3117.69, 'duration': 2.481}, {'end': 3128.173, 'text': "It's not gonna take decades, it's better a few years, optimistically, but I think that's possible.", 'start': 3121.611, 'duration': 6.562}], 'summary': 'Engineering problems to solve, detailed understanding, roadmap for solution in a few years.', 'duration': 31.459, 'max_score': 3096.714, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/Z1KwkpTUbkg/pics/Z1KwkpTUbkg3096714.jpg'}, {'end': 3300.085, 'src': 'embed', 'start': 3268.244, 'weight': 7, 'content': [{'end': 3269.985, 'text': 'the same algorithm does robotics right?', 'start': 3268.244, 'duration': 1.741}, {'end': 3272.266, 'text': 'It moves my arms, my eyes, my body right?', 'start': 3270.005, 'duration': 2.261}, {'end': 3278.137, 'text': 'And so in the future, to me, robotics and AI will merge.', 'start': 3274.715, 'duration': 3.422}, {'end': 3287.04, 'text': "They're not going to be separate fields, because the algorithms to really controlling robots are going to be the same algorithms we have in our brain,", 'start': 3278.597, 'duration': 8.443}, {'end': 3288.381, 'text': 'these sensory motor algorithms.', 'start': 3287.04, 'duration': 1.341}, {'end': 3293.943, 'text': "Today we're not there, but I think that's going to happen.", 'start': 3288.941, 'duration': 5.002}, {'end': 3296.223, 'text': 'But not all AI systems will be robotics.', 'start': 3294.083, 'duration': 2.14}, {'end': 3300.085, 'text': 'You can have systems that have very different types of embodiments.', 'start': 3297.344, 'duration': 2.741}], 'summary': 'In the future, robotics and ai will merge using sensory motor algorithms, although not all ai systems will be robotics.', 'duration': 31.841, 'max_score': 3268.244, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/Z1KwkpTUbkg/pics/Z1KwkpTUbkg3268244.jpg'}], 'start': 2213.776, 'title': 'Neural prediction mechanisms and brain engineering', 'summary': 'Discusses how predictions occur in neurons, particularly pyramidal cells, and their impact on network behavior, introduces the thousand brains theory and the necessity of reference frames, and explores the potential integration of ai and robotics with a prediction of their eventual merger.', 'chapters': [{'end': 2487.641, 'start': 2213.776, 'title': 'Neural prediction mechanisms', 'summary': 'Discusses how most predictions occur inside individual neurons, particularly the pyramidal cells, through internal dendritic spikes, which lead to altered network behavior and different representations in the brain.', 'duration': 273.865, 'highlights': ['Most predictions occur inside individual neurons, especially the pyramidal cells, through internal dendritic spikes, which are far more numerous than action potentials. This provides insight into the neural mechanism of prediction and highlights the prevalence of internal predictions in the brain.', "The emission of spikes by neurons in a predictive state leads to altered network behavior and different representations in the brain, impacting responses under different contexts. This demonstrates how internal predictions within neurons influence the overall network behavior and the brain's response to varying inputs.", 'Understanding the prediction mechanism of a single neuron can provide deep insights into the prediction capabilities of larger brains and mini brains. This emphasizes the broader implications of studying neural prediction mechanisms and their potential application to different brain structures.']}, {'end': 3022.01, 'start': 2487.781, 'title': 'Thousand brains theory and reference frames', 'summary': 'Introduces the thousand brains theory, explaining that every cortical column in the neocortex serves as a complete modeling system, resulting in thousands of models for objects like a coffee cup, and the necessity of reference frames in making predictions, highlighting the sophisticated modeling system in cortical columns and the universal application of the single cortical algorithm.', 'duration': 534.229, 'highlights': ['The thousand brains theory states that every cortical column in the neocortex is a complete modeling system, resulting in thousands of models for objects like a coffee cup. The theory proposes that the neocortex contains numerous models for objects, such as a coffee cup, demonstrating the extensive modeling capabilities of the brain.', 'The necessity of reference frames in making predictions is emphasized, as it is crucial for the cortex to know the location of objects relative to touch or vision, showcasing the importance of reference frames in cognitive processes. The transcript delves into the critical role of reference frames in predicting sensations and understanding the spatial relationships of objects, highlighting their significance in cognitive processes.', "The sophisticated modeling system in every cortical column is highlighted, revealing its capacity to learn the hierarchical structure of objects and conceptual entities, showcasing the intricate and universal modeling abilities of the brain. The discussion emphasizes the advanced modeling capabilities present in cortical columns, enabling them to comprehend the hierarchical structure of objects and conceptual entities, demonstrating the brain's sophisticated modeling abilities.", "The single cortical algorithm is mentioned as the underlying principle for representing common sense and higher-level concepts, revealing the universal application of this algorithm in the brain's cognitive processes. The transcript underscores the presence of a single cortical algorithm that underlies the representation of common sense and higher-level concepts, highlighting its universal application in the brain's cognitive processes."]}, {'end': 3322.912, 'start': 3022.03, 'title': 'Brain engineering and ai integration', 'summary': 'Discusses the brain as an engineering marvel, exploring the complexity of neural tissue and the potential for integrating ai and robotics, with a prediction that ai and robotics will eventually merge due to shared sensory motor algorithms.', 'duration': 300.882, 'highlights': ['AI and robotics will merge due to shared sensory motor algorithms. The chapter predicts that AI and robotics will merge as the algorithms controlling robots will be the same as those in the human brain, specifically the sensory motor algorithms.', "The brain is an engineering marvel with complex neural tissue. The complexity of neural tissue, including grid cells and place cells, is highlighted, emphasizing the brain's engineering complexity.", 'Understanding the brain is an engineering problem that can be solved in a few years. The understanding of the brain in detail is emphasized, with the assertion that it is an engineering problem that can be solved within a few years.']}], 'duration': 1109.136, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/Z1KwkpTUbkg/pics/Z1KwkpTUbkg2213776.jpg', 'highlights': ['Most predictions occur inside individual neurons, especially the pyramidal cells, through internal dendritic spikes, providing insight into the neural mechanism of prediction.', 'The emission of spikes by neurons in a predictive state leads to altered network behavior and different representations in the brain, impacting responses under different contexts.', 'Understanding the prediction mechanism of a single neuron can provide deep insights into the prediction capabilities of larger brains and mini brains.', 'The thousand brains theory states that every cortical column in the neocortex is a complete modeling system, resulting in thousands of models for objects like a coffee cup.', 'The necessity of reference frames in making predictions is emphasized, showcasing the importance of reference frames in cognitive processes.', 'The sophisticated modeling system in every cortical column is highlighted, revealing its capacity to learn the hierarchical structure of objects and conceptual entities.', "The single cortical algorithm is mentioned as the underlying principle for representing common sense and higher-level concepts, revealing the universal application of this algorithm in the brain's cognitive processes.", 'AI and robotics will merge due to shared sensory motor algorithms, predicting their eventual merger.', "The brain is an engineering marvel with complex neural tissue, emphasizing the brain's engineering complexity.", 'Understanding the brain is an engineering problem that can be solved in a few years, emphasizing the potential for detailed understanding within a short timeframe.']}, {'end': 4708.815, 'segs': [{'end': 3347.153, 'src': 'embed', 'start': 3323.192, 'weight': 0, 'content': [{'end': 3331.116, 'text': 'This is why I think this is I think this impact of AI is gonna be as large, if not larger, than computing has been in the last century by far,', 'start': 3323.192, 'duration': 7.924}, {'end': 3333.938, 'text': "because It's getting at a fundamental thing.", 'start': 3331.116, 'duration': 2.822}, {'end': 3335.84, 'text': "It's not a vision system or a learning system.", 'start': 3333.958, 'duration': 1.882}, {'end': 3338.383, 'text': "It's not a vision system or a hearing system.", 'start': 3335.96, 'duration': 2.423}, {'end': 3339.704, 'text': 'It is a learning system.', 'start': 3338.783, 'duration': 0.921}, {'end': 3344.189, 'text': "It's a fundamental principle, how you learn the structure in the world, how you can gain knowledge and be intelligent.", 'start': 3339.724, 'duration': 4.465}, {'end': 3347.153, 'text': "And that's what the thousand brain says was going on.", 'start': 3344.81, 'duration': 2.343}], 'summary': "Ai's impact may surpass computing in last century, as it's a fundamental learning system.", 'duration': 23.961, 'max_score': 3323.192, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/Z1KwkpTUbkg/pics/Z1KwkpTUbkg3323192.jpg'}, {'end': 3635.594, 'src': 'embed', 'start': 3608.457, 'weight': 1, 'content': [{'end': 3615.602, 'text': "Yeah, and some people decide they don't want to live, and some people decide you know you can, but the desire to live is built in DNA, right?", 'start': 3608.457, 'duration': 7.145}, {'end': 3622.467, 'text': "But I think what I'm trying to get to is in order to accomplish goals, it's useful to have the urgency of mortality.", 'start': 3615.763, 'duration': 6.704}, {'end': 3625.97, 'text': "It's what the Stoics talked about, is meditating in your mortality.", 'start': 3622.487, 'duration': 3.483}, {'end': 3635.594, 'text': 'It might be a very useful thing thing to do, to die and have the urgency of death and to realize that, to, uh,', 'start': 3626.47, 'duration': 9.124}], 'summary': 'Urgency of mortality can drive goal attainment, as stoics emphasized meditating on death.', 'duration': 27.137, 'max_score': 3608.457, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/Z1KwkpTUbkg/pics/Z1KwkpTUbkg3608457.jpg'}, {'end': 3834.84, 'src': 'embed', 'start': 3809.379, 'weight': 2, 'content': [{'end': 3814.841, 'text': "because I've studied cars quite a bit and I've studied pedestrians and cyclists quite a bit.", 'start': 3809.379, 'duration': 5.462}, {'end': 3816.882, 'text': "And there's part of me that thinks..", 'start': 3815.501, 'duration': 1.381}, {'end': 3825.268, 'text': 'that there needs to be more intelligence than we realize in order to drive successfully.', 'start': 3818.7, 'duration': 6.568}, {'end': 3834.84, 'text': 'That game theory of human interaction seems to require some deep understanding of human nature.', 'start': 3825.288, 'duration': 9.552}], 'summary': 'Studied cars, pedestrians, and cyclists. recognizes need for intelligence in driving.', 'duration': 25.461, 'max_score': 3809.379, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/Z1KwkpTUbkg/pics/Z1KwkpTUbkg3809379.jpg'}, {'end': 4006.058, 'src': 'embed', 'start': 3977.604, 'weight': 3, 'content': [{'end': 3986.188, 'text': 'So in this way, when we build AI systems, We have to make choices about how we embed them.', 'start': 3977.604, 'duration': 8.584}, {'end': 3987.628, 'text': 'So I talk about this in the book.', 'start': 3986.788, 'duration': 0.84}, {'end': 3991.41, 'text': 'I said, you know, intelligence system is not just the neocortex equivalent.', 'start': 3987.648, 'duration': 3.762}, {'end': 3996.713, 'text': 'You have to have that, but it has to have some kind of embodiment, physical or virtual.', 'start': 3991.811, 'duration': 4.902}, {'end': 3998.294, 'text': 'It has to have some sort of goals.', 'start': 3997.073, 'duration': 1.221}, {'end': 4006.058, 'text': "It has to have some sort of ideas about dangers, about things it shouldn't do, like we build in safeguards into systems.", 'start': 3998.354, 'duration': 7.704}], 'summary': 'Ai systems need embodiment, goals, and safeguards for effective functioning.', 'duration': 28.454, 'max_score': 3977.604, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/Z1KwkpTUbkg/pics/Z1KwkpTUbkg3977604.jpg'}, {'end': 4280.461, 'src': 'embed', 'start': 4251.444, 'weight': 4, 'content': [{'end': 4255.627, 'text': "Yeah Okay, so am I worried about it? Yeah, I'm totally worried about it.", 'start': 4251.444, 'duration': 4.183}, {'end': 4260.651, 'text': "The narrow component we're talking about now is the existential risk of AI.", 'start': 4256.548, 'duration': 4.103}, {'end': 4265.656, 'text': 'I want to make that distinction because I think AI can be applied poorly.', 'start': 4262.392, 'duration': 3.264}, {'end': 4270.842, 'text': 'It can be applied in ways that people are going to understand the consequences of it.', 'start': 4265.676, 'duration': 5.166}, {'end': 4275.227, 'text': 'These are all potentially very bad things.', 'start': 4272.304, 'duration': 2.923}, {'end': 4280.461, 'text': "but they're not the AI system creating this existential risk on its own.", 'start': 4276.24, 'duration': 4.221}], 'summary': 'Existential risk of ai is a concern due to potential misuse and consequences.', 'duration': 29.017, 'max_score': 4251.444, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/Z1KwkpTUbkg/pics/Z1KwkpTUbkg4251444.jpg'}, {'end': 4511.561, 'src': 'embed', 'start': 4482.241, 'weight': 5, 'content': [{'end': 4486.209, 'text': 'The point is, as a society, when we wanna look at existential risks,', 'start': 4482.241, 'duration': 3.968}, {'end': 4493.769, 'text': 'The existential risks we face that we can control almost all evolve around self-replication.', 'start': 4487.125, 'duration': 6.644}, {'end': 4501.874, 'text': "Yes The question is, I don't see a good way to make a lot of money by engineering viruses and deploying them on the world.", 'start': 4493.949, 'duration': 7.925}, {'end': 4504.496, 'text': 'There could be applications that are useful.', 'start': 4502.135, 'duration': 2.361}, {'end': 4505.697, 'text': "But let's separate out.", 'start': 4504.516, 'duration': 1.181}, {'end': 4506.938, 'text': "let's separate out.", 'start': 4505.697, 'duration': 1.241}, {'end': 4507.898, 'text': "I mean you don't need to.", 'start': 4506.938, 'duration': 0.96}, {'end': 4511.561, 'text': "you only need some terrorists who want to do it, because it doesn't take a lot of money to make viruses.", 'start': 4507.898, 'duration': 3.663}], 'summary': 'Existential risks revolve around self-replication; engineering viruses poses danger.', 'duration': 29.32, 'max_score': 4482.241, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/Z1KwkpTUbkg/pics/Z1KwkpTUbkg4482241.jpg'}, {'end': 4619.806, 'src': 'embed', 'start': 4586.867, 'weight': 6, 'content': [{'end': 4592.13, 'text': 'no, that means we regulate we, we understand the risks, we, uh, we regulate them.', 'start': 4586.867, 'duration': 5.263}, {'end': 4593.93, 'text': 'yeah, uh, you know.', 'start': 4592.13, 'duration': 1.8}, {'end': 4600.672, 'text': "look, there's a lot of things we could do a society which have some sort of financial benefit to someone, which could do a lot of harm,", 'start': 4593.93, 'duration': 6.742}, {'end': 4602.753, 'text': 'and we have to learn how to regulate those things.', 'start': 4600.672, 'duration': 2.081}, {'end': 4604.614, 'text': 'we have to learn how to deal with those things.', 'start': 4602.753, 'duration': 1.861}, {'end': 4605.374, 'text': 'i will argue this.', 'start': 4604.614, 'duration': 0.76}, {'end': 4612.178, 'text': 'i would say the opposite, like i would say having intelligent machines at our disposal will actually help us in the end more,', 'start': 4605.374, 'duration': 6.804}, {'end': 4614.04, 'text': "because it'll help us understand these risks better.", 'start': 4612.178, 'duration': 1.862}, {'end': 4615.382, 'text': "It'll help us mitigate these risks better.", 'start': 4614.06, 'duration': 1.322}, {'end': 4618.905, 'text': 'There might be ways of saying oh well, how do we solve climate change problems?', 'start': 4615.662, 'duration': 3.243}, {'end': 4619.806, 'text': 'How do we do this?', 'start': 4619.206, 'duration': 0.6}], 'summary': 'Regulating risks and embracing intelligent machines for better risk understanding and mitigation.', 'duration': 32.939, 'max_score': 4586.867, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/Z1KwkpTUbkg/pics/Z1KwkpTUbkg4586867.jpg'}], 'start': 3323.192, 'title': "Ai's societal impact", 'summary': 'Delves into the societal impact of ai, discussing fundamental principles, integration into human world, existential risks, and the need for regulation and safeguards to control behavior, highlighting concerns about automated ai building.', 'chapters': [{'end': 3769.213, 'start': 3323.192, 'title': 'Impact of ai on society', 'summary': 'Explores the potential impact of ai on society, discussing the fundamental principles of ai, the integration of intelligent systems into the human world, and the concept of consciousness in machines, challenging the notion of assigning desires and goals to intelligent machines.', 'duration': 446.021, 'highlights': ['The impact of AI on society is predicted to be as large, if not larger, than computing in the last century, due to its fundamental learning principle, creating a need to consider the integration of intelligent systems into the human world and the concept of consciousness in machines, challenging the notion of assigning desires and goals to intelligent machines.', 'Computers, being a million times faster than humans, raise the challenge of integrating them into society, which is further complicated by the question of whether intelligent machines will have the same emotions and desires as humans, leading to a discussion about the consciousness and goal-setting of intelligent machines.', 'The discussion delves into the evolutionary basis of desires and the potential role of mortality urgency in accomplishing goals, linking this to the concept of consciousness in machines and the necessity of assigning desires and goals to intelligent systems based on the applications, challenging the conventional understanding of consciousness in machines.']}, {'end': 4203.463, 'start': 3769.734, 'title': 'Intelligence and driving: a deep dive', 'summary': 'Discusses the need to explicitly teach intelligence to a smart car, the interaction between cars and pedestrians, cultural influences on human interactions, and the necessity of embedding goals and safeguards into ai systems in order to control their behavior.', 'duration': 433.729, 'highlights': ['The interaction between cars and pedestrians requires a deep understanding of human nature, involving a mutual dance of trust and potential danger, which may need to be explicitly taught to an intelligent machine. The interaction between cars and pedestrians involves a mutual dance of trust and potential danger, which may need to be explicitly taught to an intelligent machine to successfully operate in real-world scenarios.', 'Cultural influences impact pedestrian interactions, suggesting that intelligence systems could learn and understand human behaviors, although this does not necessarily mean they will develop their own personal agency or goals. Cultural influences impact pedestrian interactions, suggesting that intelligence systems could learn and understand human behaviors, although this does not necessarily mean they will develop their own personal agency or goals.', 'The necessity of embedding goals and safeguards into AI systems is emphasized to control their behavior, as intelligence itself does not inherently lead to the development of personal agency or goals. The necessity of embedding goals and safeguards into AI systems is emphasized to control their behavior, as intelligence itself does not inherently lead to the development of personal agency or goals.', "The speaker's confidence in understanding the mechanisms of AI creation, emphasizing the importance of detailed engineering and the absence of automatic evolution of personal agency in intelligent systems. The speaker expresses confidence in understanding the mechanisms of AI creation, emphasizing the importance of detailed engineering and the absence of automatic evolution of personal agency in intelligent systems."]}, {'end': 4412.921, 'start': 4203.483, 'title': "Ai's existential risk", 'summary': "Discusses the potential societal impact and existential risk of ai, emphasizing the distinction between ai's application and its autonomous existential risk, while expressing concern about the impact of ai at scale and the possibility of automated ai building.", 'duration': 209.438, 'highlights': ["The chapter discusses the potential societal impact and existential risk of AI, emphasizing the distinction between AI's application and its autonomous existential risk. The speaker raises concerns about how the application of recommender systems at scale can change societies and expresses worry about the existential risk of AI, emphasizing the distinction between AI's application and its autonomous existential risk.", 'The impact of AI at scale and the possibility of automated AI building. The speaker highlights the potential societal impact of AI at scale and raises the possibility of automating AI building, discussing the efficiency and automation of factories as a means to create robots with significant societal benefits.', "AI's potential to not inherently pose an existential risk on its own. The speaker expresses the belief that AI systems are unlikely to autonomously pose an existential risk and underscores the difficulty of AI systems trying to kill off the human race, stating that AI is not inherently predisposed to such actions."]}, {'end': 4708.815, 'start': 4413.061, 'title': 'Risks of self-replication in ai', 'summary': 'Discusses the potential dangers of self-replicating systems in ai, emphasizing the need for regulation to mitigate existential risks and the societal impact of intelligent machines.', 'duration': 295.754, 'highlights': ['The existential risks we face that we can control almost all evolve around self-replication. Emphasizes the centrality of self-replication as the primary existential risk in AI.', 'Creating ultra-intelligent, super-intelligent systems is not necessarily coupled with self-replicating systems. Dispels the notion that intelligence and self-replication are inherently linked, highlighting the independence of creating intelligent systems from self-replication.', 'Regulation and understanding of the risks are essential in managing the development of intelligent systems. Stresses the importance of regulation and risk understanding to mitigate potential harm from the development of intelligent systems.', 'The more intelligent systems we build, the more tempting it becomes to create self-reproducing systems. Raises awareness of the correlation between the advancement of intelligent systems and the temptation to create self-replicating systems.', 'The potential societal benefits of having intelligent machines at our disposal in understanding and mitigating risks such as climate change. Highlights the potential positive impact of intelligent machines in addressing societal challenges like climate change.']}], 'duration': 1385.623, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/Z1KwkpTUbkg/pics/Z1KwkpTUbkg3323192.jpg', 'highlights': ['The impact of AI on society is predicted to be as large, if not larger, than computing in the last century, due to its fundamental learning principle, creating a need to consider the integration of intelligent systems into the human world and the concept of consciousness in machines, challenging the notion of assigning desires and goals to intelligent machines.', 'The discussion delves into the evolutionary basis of desires and the potential role of mortality urgency in accomplishing goals, linking this to the concept of consciousness in machines and the necessity of assigning desires and goals to intelligent systems based on the applications, challenging the conventional understanding of consciousness in machines.', 'The interaction between cars and pedestrians requires a deep understanding of human nature, involving a mutual dance of trust and potential danger, which may need to be explicitly taught to an intelligent machine.', 'The necessity of embedding goals and safeguards into AI systems is emphasized to control their behavior, as intelligence itself does not inherently lead to the development of personal agency or goals.', "The chapter discusses the potential societal impact and existential risk of AI, emphasizing the distinction between AI's application and its autonomous existential risk.", 'The existential risks we face that we can control almost all evolve around self-replication. Emphasizes the centrality of self-replication as the primary existential risk in AI.', 'Regulation and understanding of the risks are essential in managing the development of intelligent systems. Stresses the importance of regulation and risk understanding to mitigate potential harm from the development of intelligent systems.', 'The potential societal benefits of having intelligent machines at our disposal in understanding and mitigating risks such as climate change. Highlights the potential positive impact of intelligent machines in addressing societal challenges like climate change.']}, {'end': 6244.083, 'segs': [{'end': 4866.612, 'src': 'embed', 'start': 4837.906, 'weight': 0, 'content': [{'end': 4842.228, 'text': 'Is it one of the dreams is to upload our minds in the digital space?', 'start': 4837.906, 'duration': 4.322}, {'end': 4848.71, 'text': 'So can we just give our minds to these systems so they can operate on them?', 'start': 4842.408, 'duration': 6.302}, {'end': 4851.871, 'text': 'Is there some kind of more interesting merger or is there more??', 'start': 4849.01, 'duration': 2.861}, {'end': 4852.992, 'text': 'And the third part of my book?', 'start': 4852.051, 'duration': 0.941}, {'end': 4855.652, 'text': 'I talked about all these scenarios, and let me just walk through them.', 'start': 4852.992, 'duration': 2.66}, {'end': 4859.727, 'text': 'Sure The uploading the mind one.', 'start': 4855.773, 'duration': 3.954}, {'end': 4863.169, 'text': 'Yes Extremely, really difficult to do.', 'start': 4860.027, 'duration': 3.142}, {'end': 4866.612, 'text': 'We have no idea how to do this, even remotely right now.', 'start': 4864.03, 'duration': 2.582}], 'summary': 'Exploring the challenges of uploading minds into digital space.', 'duration': 28.706, 'max_score': 4837.906, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/Z1KwkpTUbkg/pics/Z1KwkpTUbkg4837906.jpg'}, {'end': 4990.486, 'src': 'embed', 'start': 4966.198, 'weight': 1, 'content': [{'end': 4973.683, 'text': 'The merging your mind with the neural link thing, right? Again, really, really difficult.', 'start': 4966.198, 'duration': 7.485}, {'end': 4977.024, 'text': "It's one thing to make progress to control a prosthetic arm.", 'start': 4974.043, 'duration': 2.981}, {'end': 4982.345, 'text': "It's another to have like a billion or several billion things and understanding what those signals mean.", 'start': 4977.084, 'duration': 5.261}, {'end': 4987.126, 'text': "It's the one thing that like, okay, I can learn to think some patterns to make something happen.", 'start': 4983.125, 'duration': 4.001}, {'end': 4990.486, 'text': "It's quite another thing to have a system, a computer,", 'start': 4987.726, 'duration': 2.76}], 'summary': 'Merging mind with neural link is challenging due to complexity of interpreting signals and controlling numerous things.', 'duration': 24.288, 'max_score': 4966.198, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/Z1KwkpTUbkg/pics/Z1KwkpTUbkg4966198.jpg'}, {'end': 5708.629, 'src': 'embed', 'start': 5679.767, 'weight': 2, 'content': [{'end': 5682.669, 'text': 'So one very simple thing I said how would we archive what we know?', 'start': 5679.767, 'duration': 2.902}, {'end': 5684.17, 'text': 'That was a very simple idea.', 'start': 5683.149, 'duration': 1.021}, {'end': 5686.231, 'text': "I said, you know what that wouldn't be that hard to put.", 'start': 5684.19, 'duration': 2.041}, {'end': 5691.553, 'text': "a few satellites, you know, going around the sun and we'd upload Wikipedia every day and that kind of thing.", 'start': 5686.231, 'duration': 5.322}, {'end': 5697.036, 'text': "So, you know, if we end up killing ourselves, well, it's up there and the next intelligence piece will find it and learn something.", 'start': 5692.434, 'duration': 4.602}, {'end': 5697.897, 'text': 'They would like that.', 'start': 5697.296, 'duration': 0.601}, {'end': 5698.757, 'text': 'They would appreciate that.', 'start': 5698.057, 'duration': 0.7}, {'end': 5702.079, 'text': "So that's one thing.", 'start': 5701.398, 'duration': 0.681}, {'end': 5708.629, 'text': 'The next thing I said, well, what if, you know, Outside of our solar system, we have the SETI program.', 'start': 5702.159, 'duration': 6.47}], 'summary': 'Proposing to archive wikipedia on satellites for future intelligence to access.', 'duration': 28.862, 'max_score': 5679.767, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/Z1KwkpTUbkg/pics/Z1KwkpTUbkg5679767.jpg'}, {'end': 5847.392, 'src': 'embed', 'start': 5818.591, 'weight': 3, 'content': [{'end': 5820.972, 'text': 'you ought to figure out how would we create a signal?', 'start': 5818.591, 'duration': 2.381}, {'end': 5822.413, 'text': 'Like, what would we create?', 'start': 5821.292, 'duration': 1.121}, {'end': 5828.516, 'text': 'that would be like that that would persist for millions of years, that would be broadcast broadly, that you could see from a distance.', 'start': 5822.413, 'duration': 6.103}, {'end': 5832.438, 'text': 'that was unequivocal, came from an intelligent species.', 'start': 5828.516, 'duration': 3.922}, {'end': 5836.14, 'text': "And so I gave that one example, because they don't know what I know of, actually.", 'start': 5833.158, 'duration': 2.982}, {'end': 5847.392, 'text': 'And then finally, right, If ultimately our solar system will die at some point in time, how do we go beyond that?', 'start': 5836.68, 'duration': 10.712}], 'summary': 'Creating a signal to persist for millions of years, broadcasting broadly to signify an intelligent species and addressing the longevity of our solar system.', 'duration': 28.801, 'max_score': 5818.591, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/Z1KwkpTUbkg/pics/Z1KwkpTUbkg5818591.jpg'}, {'end': 6066.548, 'src': 'embed', 'start': 6039.125, 'weight': 4, 'content': [{'end': 6042.522, 'text': "yeah, Understanding how the brain works, we're going to figure that out once.", 'start': 6039.125, 'duration': 3.397}, {'end': 6043.122, 'text': "That's it.", 'start': 6042.782, 'duration': 0.34}, {'end': 6044.122, 'text': "It's going to be figured out once.", 'start': 6043.242, 'duration': 0.88}, {'end': 6046.683, 'text': "And after that, that's the answer.", 'start': 6044.762, 'duration': 1.921}, {'end': 6050.164, 'text': 'And people will study that thousands of years from now.', 'start': 6046.763, 'duration': 3.401}, {'end': 6054.085, 'text': 'We still venerate Newton and Einstein.', 'start': 6050.744, 'duration': 3.341}, {'end': 6061.486, 'text': 'Because ideas are exciting even well into the future.', 'start': 6056.385, 'duration': 5.101}, {'end': 6066.548, 'text': "Well, the interesting thing is big ideas, even if they're wrong, are still useful.", 'start': 6061.546, 'duration': 5.002}], 'summary': "Understanding the brain's workings will be figured out once, and will be studied for thousands of years. big ideas, even if wrong, remain useful.", 'duration': 27.423, 'max_score': 6039.125, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/Z1KwkpTUbkg/pics/Z1KwkpTUbkg6039125.jpg'}], 'start': 4708.855, 'title': 'Future integration of ai and human consciousness', 'summary': 'Discusses the challenges and implications of merging human consciousness with ai, exploring the complexity of defining human identity, preserving knowledge for future civilizations, and the potential of brain-computer merging, providing insights into the future of humanity and communication.', 'chapters': [{'end': 4944.917, 'start': 4708.855, 'title': 'Future of ai and human integration', 'summary': 'Discusses the implications of self-replicating systems and the challenges of merging human minds with super intelligent cortical systems, highlighting the impracticality of uploading human minds into digital space and the complexity of defining human identity.', 'duration': 236.062, 'highlights': ['The challenges of uploading human minds into digital space and the impracticality of the process are discussed, pointing out the difficulty in replicating the entire body, not just the brain, and the potential dissatisfaction with the result.', 'The concept of self-replication and the role of humans in the loop are explored, emphasizing the limitations of true self-replication in the presence of human intervention and decision-making.', 'The potential implications of merging with super intelligent cortical systems are considered, raising questions about the future coexistence of humans and intelligent beings and the uncertainty surrounding the preservation of human identity in such scenarios.']}, {'end': 5413.27, 'start': 4945.017, 'title': 'Challenges of merging human consciousness with ai', 'summary': 'Discusses the daunting challenges of merging human consciousness with ai, highlighting the difficulty of uploading the brain and understanding and integrating billions of neural signals, as well as exploring the potential transcendent role of intelligent machines in overcoming human biological limitations and propelling human knowledge throughout the universe.', 'duration': 468.253, 'highlights': ['Understanding and integrating billions of neural signals is a major challenge in merging human consciousness with AI, as it involves a level of complexity that exceeds current capabilities.', 'The difficulty of uploading the brain is emphasized, with a timeframe of potentially a thousand years or more before this becomes feasible due to both biological and machine learning obstacles.', 'The potential transcendent role of intelligent machines is explored, offering the possibility of overcoming human biological limitations and propelling human knowledge throughout the universe, including the construction and terraforming of distant habitats.', 'The chapter also delves into the limitations of human biology, the potential for self-destruction, and the potential for intelligent machines to represent and propagate human knowledge and history throughout the universe.', 'The discussion also touches on the marginal benefits of merging with AI, potential applications such as care robots and smarter cars, and the broader implications for humanity as a whole.']}, {'end': 5796.444, 'start': 5413.53, 'title': 'Preserving human knowledge for the future', 'summary': 'Discusses the importance of preserving human knowledge and the potential extinction of intelligent civilizations, emphasizing the need to archive our knowledge and create long-lasting signals for future intelligent life, while also acknowledging the possibility of previous intelligent civilizations on earth and in our galaxy.', 'duration': 382.914, 'highlights': ['The importance of archiving human knowledge is emphasized, suggesting the idea of uploading Wikipedia to satellites orbiting the sun to ensure preservation for future intelligent life. Importance of archiving human knowledge, idea of uploading Wikipedia to satellites, preservation for future intelligent life', 'Consideration is given to the potential extinction of intelligent civilizations, indicating the need for long-lasting signals that could be detected by future civilizations, with a proposed concept involving the creation of unique patterns around the sun to signify the existence of previous intelligent life. Concern for potential extinction of intelligent civilizations, need for long-lasting signals, proposed concept of creating unique patterns around the sun', 'Discussion on the possibility of previous intelligent civilizations on Earth and in the galaxy, contemplating the potential disappearance of signals from these civilizations and the challenges of detecting such long-lasting signals through programs like SETI. Possibility of previous intelligent civilizations, challenges of detecting long-lasting signals, implications for programs like SETI']}, {'end': 5997.783, 'start': 5798.005, 'title': 'The future of humanity and communication', 'summary': 'Discusses the potential for continuous broadcasting of signals from intelligent species, the need to preserve humanity for future civilizations, and the transcendence of ideas beyond our time and planet, providing insights into the future of humanity and communication.', 'duration': 199.778, 'highlights': ['The potential for continuous broadcasting of signals from intelligent species The speaker argues for the SETI program to look for signals that are broadcast broadly, persist for millions of years, and are unequivocal, indicating an intelligent species.', 'Preserving humanity for future civilizations The speaker reflects on the concept of preserving oneself for future civilizations and discusses the relevance of ideas and communication transcending time, aiming to make his book relevant to readers 100 years from now.', 'The transcendence of ideas beyond our time and planet The speaker emphasizes the importance of discussing ideas that could transcend a thousand or million years, and reflects on making his book and discussions relevant to future generations and potential alien readers.']}, {'end': 6244.083, 'start': 5997.783, 'title': 'Future of brain-computer merging', 'summary': 'Discusses the potential of merging brains with computers and the understanding of complex systems, expressing optimism in achieving complete understanding of the human brain and acknowledging the potential limitations of current ideas and models.', 'duration': 246.3, 'highlights': ['Understanding how the brain works will be a one-time achievement, impacting the future for thousands of years.', 'Big ideas, even if partially wrong, can still be useful, as demonstrated by the examples of Newton and Einstein.', 'Expressing optimism in completely understanding the human brain and comparing it to the understanding of complex systems and molecules.', 'Acknowledging the potential limitations and uncertainties in current ideas and models, particularly in the context of voting and representations.']}], 'duration': 1535.228, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/Z1KwkpTUbkg/pics/Z1KwkpTUbkg4708855.jpg', 'highlights': ['The challenges of uploading human minds into digital space and the impracticality of replicating the entire body are discussed.', 'Understanding and integrating billions of neural signals is a major challenge in merging human consciousness with AI.', 'The importance of archiving human knowledge is emphasized, suggesting the idea of uploading Wikipedia to satellites orbiting the sun.', 'The potential for continuous broadcasting of signals from intelligent species is discussed.', 'Understanding how the brain works will be a one-time achievement, impacting the future for thousands of years.']}, {'end': 7100.404, 'segs': [{'end': 6384.506, 'src': 'embed', 'start': 6334.178, 'weight': 1, 'content': [{'end': 6336.279, 'text': "But that's the role of theorists play.", 'start': 6334.178, 'duration': 2.101}, {'end': 6344.382, 'text': 'They give you a framework on which you now can talk about a problem and figure out, OK, now we can start digging more details.', 'start': 6336.319, 'duration': 8.063}, {'end': 6348.604, 'text': 'The best frameworks stick around while the details change.', 'start': 6344.782, 'duration': 3.822}, {'end': 6357.864, 'text': "Again, the classic example is Newton and Einstein, right? Newton's theories are still used.", 'start': 6348.624, 'duration': 9.24}, {'end': 6359.146, 'text': "They're still valuable.", 'start': 6358.024, 'duration': 1.122}, {'end': 6359.986, 'text': "They're still practical.", 'start': 6359.186, 'duration': 0.8}, {'end': 6361.088, 'text': "They're not like wrong.", 'start': 6360.006, 'duration': 1.082}, {'end': 6362.729, 'text': "It's just they've been refined.", 'start': 6361.108, 'duration': 1.621}, {'end': 6363.991, 'text': "Yeah, but that's in physics.", 'start': 6362.749, 'duration': 1.242}, {'end': 6365.692, 'text': "It's not obvious, by the way.", 'start': 6364.071, 'duration': 1.621}, {'end': 6371.658, 'text': "It's not obvious for physics either that the universe should be such that it's amenable to these simple..", 'start': 6366.053, 'duration': 5.605}, {'end': 6373.2, 'text': 'I know, but so far it appears to be.', 'start': 6371.658, 'duration': 1.542}, {'end': 6375.362, 'text': 'As far as we can tell.', 'start': 6374.421, 'duration': 0.941}, {'end': 6379.684, 'text': 'Yeah, I mean, but as far as we could tell.', 'start': 6375.382, 'duration': 4.302}, {'end': 6384.506, 'text': "But it's also an open question whether the brain is amenable to such clean theories.", 'start': 6380.264, 'duration': 4.242}], 'summary': "The role of theorists is to provide lasting frameworks for problem-solving, as seen in physics with newton's theories still being used.", 'duration': 50.328, 'max_score': 6334.178, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/Z1KwkpTUbkg/pics/Z1KwkpTUbkg6334178.jpg'}, {'end': 6441.706, 'src': 'embed', 'start': 6410.358, 'weight': 0, 'content': [{'end': 6411.498, 'text': 'And unbelievably so.', 'start': 6410.358, 'duration': 1.14}, {'end': 6415.8, 'text': "And then I mentioned earlier, the other thing is it's universal abilities.", 'start': 6411.698, 'duration': 4.102}, {'end': 6419.744, 'text': 'It is so flexible to learn so many things.', 'start': 6416.3, 'duration': 3.444}, {'end': 6421.847, 'text': "We haven't figured out what it can't learn yet.", 'start': 6419.804, 'duration': 2.043}, {'end': 6423.429, 'text': "We don't know, but we haven't figured out yet.", 'start': 6422.007, 'duration': 1.422}, {'end': 6425.632, 'text': 'But it can learn things that it never was evolved to learn.', 'start': 6423.449, 'duration': 2.183}, {'end': 6427.094, 'text': 'So those give us hope.', 'start': 6426.232, 'duration': 0.862}, {'end': 6435.663, 'text': "That's why I went into this field, because I said, you know, This regular structure, it's doing this amazing number of things.", 'start': 6428.495, 'duration': 7.168}, {'end': 6441.706, 'text': "There's got to be some underlying principles that are common, and other scientists have come up with the same conclusions.", 'start': 6435.703, 'duration': 6.003}], 'summary': 'Ai demonstrates universal learning abilities, offering hope and potential for uncovering common underlying principles.', 'duration': 31.348, 'max_score': 6410.358, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/Z1KwkpTUbkg/pics/Z1KwkpTUbkg6410358.jpg'}, {'end': 6524.653, 'src': 'embed', 'start': 6494.181, 'weight': 3, 'content': [{'end': 6496.104, 'text': "I'm one, so I'm disappointed in myself too.", 'start': 6494.181, 'duration': 1.923}, {'end': 6499.529, 'text': "It's kind of a sad state.", 'start': 6498.007, 'duration': 1.522}, {'end': 6501.772, 'text': "There's two things that disappoint me.", 'start': 6499.549, 'duration': 2.223}, {'end': 6513.049, 'text': "One is how it's difficult for us to separate our rational component of ourselves from our evolutionary heritage, which is, you know,", 'start': 6502.954, 'duration': 10.095}, {'end': 6514.729, 'text': 'not always pretty, you know.', 'start': 6513.049, 'duration': 1.68}, {'end': 6521.472, 'text': 'um, rape is a is an evolutionary good strategy for reproduction.', 'start': 6514.729, 'duration': 6.743}, {'end': 6524.653, 'text': 'murder can be, at times too, you know.', 'start': 6521.472, 'duration': 3.181}], 'summary': 'Difficulty separating rationality from evolutionary heritage; mentions rape and murder as evolutionary strategies.', 'duration': 30.472, 'max_score': 6494.181, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/Z1KwkpTUbkg/pics/Z1KwkpTUbkg6494181.jpg'}, {'end': 6626.982, 'src': 'embed', 'start': 6601.608, 'weight': 4, 'content': [{'end': 6606.534, 'text': "And we can test it all the time, right? I touch something, I look at it, I talk to you, see if my model's correct.", 'start': 6601.608, 'duration': 4.926}, {'end': 6610.479, 'text': "But so much of what we know is stuff I can't directly interact with.", 'start': 6606.935, 'duration': 3.544}, {'end': 6612.181, 'text': 'I only know because someone told me about it.', 'start': 6610.519, 'duration': 1.662}, {'end': 6621.136, 'text': "so so we're prone, inherently prone, to having false beliefs, because if i'm told something, how am i going to know it's right or wrong?", 'start': 6613.107, 'duration': 8.029}, {'end': 6626.982, 'text': 'right? and so then we have the scientific process which says we are inherently flawed.', 'start': 6621.136, 'duration': 5.846}], 'summary': 'Scientific process acknowledges human susceptibility to false beliefs and flawed knowledge.', 'duration': 25.374, 'max_score': 6601.608, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/Z1KwkpTUbkg/pics/Z1KwkpTUbkg6601608.jpg'}, {'end': 6716.485, 'src': 'embed', 'start': 6694.536, 'weight': 5, 'content': [{'end': 6703.362, 'text': 'Now, this could be small-scale censorship, like a young grad student who comes up, who raises their hand and says some crazy idea.', 'start': 6694.536, 'duration': 8.826}, {'end': 6713.404, 'text': "A form of censorship could be, I shouldn't use the word censorship, but like de-incentivize them from, no, no, no, no, this is the way it's been done.", 'start': 6704.762, 'duration': 8.642}, {'end': 6715.285, 'text': "Yeah, yeah, you're a foolish kid, don't think the same.", 'start': 6713.444, 'duration': 1.841}, {'end': 6716.485, 'text': "Yeah, you're foolish.", 'start': 6715.305, 'duration': 1.18}], 'summary': 'Discussion on small-scale censorship and discouraging innovative ideas.', 'duration': 21.949, 'max_score': 6694.536, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/Z1KwkpTUbkg/pics/Z1KwkpTUbkg6694536.jpg'}, {'end': 6763.474, 'src': 'embed', 'start': 6740.923, 'weight': 6, 'content': [{'end': 6748.431, 'text': 'i know how to deal with this issue that you bring up is, if everybody understood, as part of your upbringing in life,', 'start': 6740.923, 'duration': 7.508}, {'end': 6755.382, 'text': 'something about how your brain works, that it builds a model of the world, how it works, how basically it builds that model of the world,', 'start': 6748.431, 'duration': 6.951}, {'end': 6757.585, 'text': 'and that the model is not the real world.', 'start': 6755.382, 'duration': 2.203}, {'end': 6758.306, 'text': "It's just a model.", 'start': 6757.645, 'duration': 0.661}, {'end': 6761.031, 'text': "And it's never going to reflect the entire world.", 'start': 6759.168, 'duration': 1.863}, {'end': 6761.972, 'text': 'And it can be wrong.', 'start': 6761.171, 'duration': 0.801}, {'end': 6763.474, 'text': "And it's easy to be wrong.", 'start': 6762.132, 'duration': 1.342}], 'summary': 'Our brains create models of the world, which may be incorrect, leading to potential misunderstandings.', 'duration': 22.551, 'max_score': 6740.923, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/Z1KwkpTUbkg/pics/Z1KwkpTUbkg6740923.jpg'}], 'start': 6244.203, 'title': 'Complexity and existential risks', 'summary': "Discusses the brain's complexity and the role of theorists in providing frameworks, emphasizing its regular structure and universal learning abilities. it also delves into the existential risks of human nature, including susceptibility to false beliefs and the potential consequences of censorship, while highlighting the importance of understanding the limitations of the human brain in modeling reality and pursuing knowledge.", 'chapters': [{'end': 6464.629, 'start': 6244.203, 'title': 'Complexity and theoretical frameworks', 'summary': 'Discusses the complexity of the brain and the role of theorists in providing frameworks for understanding, suggesting that while the brain is extremely messy and complex, it exhibits regular structure and universal learning abilities, providing hope for underlying common principles.', 'duration': 220.426, 'highlights': ['The brain is extremely messy and complex but exhibits regular structure and universal learning abilities. The human brain is both extremely messy and complex, yet it possesses extremely regular structure in the neocortex and universal learning abilities, providing hope for common underlying principles.', "The role of theorists in providing frameworks for understanding complex systems. Theorists play a crucial role in providing frameworks to understand complex systems, allowing for the exploration of details within a given framework, as exemplified by the enduring value of Newton's theories in physics.", 'The idea that the brain may not be amenable to simple, clean theories. It is an open question whether the brain is amenable to simple, clean theories, challenging the notion that the brain can be fully understood through simple explanations.']}, {'end': 7100.404, 'start': 6465.669, 'title': 'Human nature and existential risks', 'summary': 'Discusses the existential risks of human nature, including the difficulty of overcoming evolutionary heritage, inherent susceptibility to false beliefs, and the potential consequences of censorship, while emphasizing the importance of understanding the limitations of the human brain in modeling reality and the pursuit of knowledge.', 'duration': 634.735, 'highlights': ["The difficulty of overcoming evolutionary heritage The speaker expresses disappointment in humanity's struggle to separate the rational component from evolutionary heritage, highlighting the evolutionary strategies of rape, murder, and causing misery as concerning aspects of human nature.", "Inherent susceptibility to false beliefs The discussion emphasizes humans' inherent susceptibility to false beliefs due to the limitations in directly interacting with much of the knowledge, leading to a tendency to rely on information provided by others and the necessity of seeking contrary evidence through the scientific process.", "Consequences of censorship and de-incentivizing diverse ideas The speaker discusses the potential dangers of censorship and de-incentivizing diverse ideas, highlighting the historical significance of revolutionary ideas and advocating for skepticism and understanding of the brain's modeling process as a solution to promote a better world.", "Understanding the limitations of the human brain in modeling reality and pursuing knowledge The chapter emphasizes the importance of understanding the human brain's limitations in constructing models of the world and the necessity of skepticism towards beliefs, while acknowledging the pursuit of knowledge for its own pleasure and the unpredictable impact of acquired knowledge."]}], 'duration': 856.201, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/Z1KwkpTUbkg/pics/Z1KwkpTUbkg6244203.jpg', 'highlights': ['The human brain exhibits regular structure and universal learning abilities, providing hope for common underlying principles.', "Theorists play a crucial role in providing frameworks to understand complex systems, exemplified by the enduring value of Newton's theories in physics.", "The brain's amenability to simple, clean theories is an open question, challenging the notion of full understanding through simple explanations.", 'Humanity struggles to separate the rational component from evolutionary heritage, highlighting concerning aspects of human nature.', 'Humans have inherent susceptibility to false beliefs due to limitations in directly interacting with knowledge, necessitating seeking contrary evidence through the scientific process.', "Censorship and de-incentivizing diverse ideas pose potential dangers, advocating for skepticism and understanding of the brain's modeling process as a solution.", 'Understanding the limitations of the human brain in modeling reality and pursuing knowledge is emphasized, acknowledging the unpredictable impact of acquired knowledge.']}, {'end': 8299.874, 'segs': [{'end': 7129.573, 'src': 'embed', 'start': 7100.404, 'weight': 2, 'content': [{'end': 7104.508, 'text': "so today's airplanes fly on the same principles as birds, which is observed by.", 'start': 7100.404, 'duration': 4.104}, {'end': 7106.33, 'text': 'so everyone get that analogy wrong.', 'start': 7104.508, 'duration': 1.822}, {'end': 7108.552, 'text': "but let's step back from that right?", 'start': 7106.33, 'duration': 2.222}, {'end': 7113.858, 'text': 'once you understand the principles of flight, you can choose how to implement them.', 'start': 7108.552, 'duration': 5.306}, {'end': 7120.825, 'text': "yeah, No one's going to use bones and feathers and muscles, but they do have wings and we don't flap them.", 'start': 7113.858, 'duration': 6.967}, {'end': 7121.445, 'text': 'We have propellers.', 'start': 7120.845, 'duration': 0.6}, {'end': 7129.573, 'text': 'So when we have the principles of computation that goes on to modeling the world in a brain, we understand those principles very clearly.', 'start': 7121.465, 'duration': 8.108}], 'summary': "Airplanes follow principles of birds' flight, using propellers instead of wings. understanding flight principles leads to effective implementation.", 'duration': 29.169, 'max_score': 7100.404, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/Z1KwkpTUbkg/pics/Z1KwkpTUbkg7100404.jpg'}, {'end': 7319.715, 'src': 'embed', 'start': 7292.25, 'weight': 1, 'content': [{'end': 7298.095, 'text': "And by adding dendrites to them at just one more level of complexity, that's in biological systems,", 'start': 7292.25, 'duration': 5.845}, {'end': 7302.779, 'text': 'you can solve problems in continuous learning and rapid learning.', 'start': 7298.095, 'duration': 4.684}, {'end': 7308.816, 'text': "So we're trying to bring the existing field We'll see if we can do it.", 'start': 7303.32, 'duration': 5.496}, {'end': 7313.544, 'text': "We're trying to bring the existing field of machine learning commercially along with us.", 'start': 7308.876, 'duration': 4.668}, {'end': 7315.648, 'text': 'You brought up this idea of paying for it.', 'start': 7313.584, 'duration': 2.064}, {'end': 7319.715, 'text': 'Commercially along with us as we move towards the ultimate goal of a true AI system.', 'start': 7316.189, 'duration': 3.526}], 'summary': 'Adding dendrites in biological systems enables continuous and rapid learning, aiming to bring machine learning commercially.', 'duration': 27.465, 'max_score': 7292.25, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/Z1KwkpTUbkg/pics/Z1KwkpTUbkg7292250.jpg'}, {'end': 7373.502, 'src': 'embed', 'start': 7344.821, 'weight': 0, 'content': [{'end': 7346.883, 'text': 'Or more robust, even just more robust.', 'start': 7344.821, 'duration': 2.062}, {'end': 7353.891, 'text': 'And making it somehow much better, incentivizing sparsity somehow.', 'start': 7347.744, 'duration': 6.147}, {'end': 7358.216, 'text': "Yeah Well, if you can make things 100 times faster, then there's plenty of incentive.", 'start': 7354.052, 'duration': 4.164}, {'end': 7359.978, 'text': "That's true.", 'start': 7358.877, 'duration': 1.101}, {'end': 7365.585, 'text': 'People are spending millions of dollars just training some of these networks now, these transformer networks.', 'start': 7360.339, 'duration': 5.246}, {'end': 7373.502, 'text': 'Let me ask you a big question for young people listening to this today in high school and college.', 'start': 7367.079, 'duration': 6.423}], 'summary': 'Improving speed by 100x can incentivize better performance in training transformer networks.', 'duration': 28.681, 'max_score': 7344.821, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/Z1KwkpTUbkg/pics/Z1KwkpTUbkg7344821.jpg'}, {'end': 7647.679, 'src': 'heatmap', 'start': 7559.054, 'weight': 1, 'content': [{'end': 7568.46, 'text': 'But then when I was 22, And I was reading the, no, excuse me, it was 1979, excuse me, 1979.', 'start': 7559.054, 'duration': 9.406}, {'end': 7571.562, 'text': 'I was reading, so I was, at that time I was 22.', 'start': 7568.46, 'duration': 3.102}, {'end': 7575.605, 'text': 'I was reading the September issue of Scientific American, which is all about the brain.', 'start': 7571.562, 'duration': 4.043}, {'end': 7581.629, 'text': 'And then the final essay was by Francis Crick, who of DNA fame.', 'start': 7576.645, 'duration': 4.984}, {'end': 7584.991, 'text': 'And he had taken his interest to studying the brain now.', 'start': 7582.569, 'duration': 2.422}, {'end': 7588.533, 'text': "And he said, you know, there's something wrong here.", 'start': 7585.611, 'duration': 2.922}, {'end': 7590.595, 'text': 'He says, we got all this data.', 'start': 7588.773, 'duration': 1.822}, {'end': 7593.827, 'text': 'Oh, this fact, this is 1979.', 'start': 7591.926, 'duration': 1.901}, {'end': 7596.508, 'text': 'All these facts about the brain, tons and tons of facts about the brain.', 'start': 7593.827, 'duration': 2.681}, {'end': 7601.77, 'text': 'Do we need more facts or do we just need to think about a way of rearranging the facts we have??', 'start': 7597.709, 'duration': 4.061}, {'end': 7604.251, 'text': "Maybe we're just not thinking about the problem correctly.", 'start': 7601.83, 'duration': 2.421}, {'end': 7609.594, 'text': "You know, if he says this shouldn't be, this shouldn't be like this, you know?", 'start': 7604.932, 'duration': 4.662}, {'end': 7616.757, 'text': "So I read that and I said wow, I said I don't have to become like an experimental neuroscientist.", 'start': 7610.314, 'duration': 6.443}, {'end': 7619.138, 'text': 'I could just look at all those facts.', 'start': 7617.077, 'duration': 2.061}, {'end': 7622.829, 'text': 'and try and become a theoretician and try to figure it out.', 'start': 7620.066, 'duration': 2.763}, {'end': 7626.854, 'text': 'And I said that I felt like it was something I would be good at.', 'start': 7623.83, 'duration': 3.024}, {'end': 7628.816, 'text': "I said, I wouldn't be a good experimentalist.", 'start': 7627.194, 'duration': 1.622}, {'end': 7633.521, 'text': "I don't have the patience for it, but I'm a good thinker and I love puzzles.", 'start': 7628.836, 'duration': 4.685}, {'end': 7635.744, 'text': 'And this is like the biggest puzzle in the world.', 'start': 7633.962, 'duration': 1.782}, {'end': 7637.285, 'text': "It's the biggest puzzle of all time.", 'start': 7635.764, 'duration': 1.521}, {'end': 7639.388, 'text': 'And I got all the puzzle pieces in front of me.', 'start': 7637.305, 'duration': 2.083}, {'end': 7641.35, 'text': 'Damn, that was exciting.', 'start': 7640.189, 'duration': 1.161}, {'end': 7647.679, 'text': "And there's something obviously you can't convert into words that just kind of sparked this passion.", 'start': 7641.636, 'duration': 6.043}], 'summary': 'Reading scientific american in 1979 sparked a passion for studying the brain and becoming a theoretician instead of an experimental neuroscientist.', 'duration': 88.625, 'max_score': 7559.054, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/Z1KwkpTUbkg/pics/Z1KwkpTUbkg7559054.jpg'}, {'end': 7942.361, 'src': 'embed', 'start': 7912.873, 'weight': 3, 'content': [{'end': 7920.708, 'text': "um uh, it doesn't impact how i think about how we model the world, But from a humanity point of view, I think it's essential.", 'start': 7912.873, 'duration': 7.835}, {'end': 7929.993, 'text': 'Well, I tend to give it to the new brain and also I tend to think that some aspects of that need to be engineered into AI systems,', 'start': 7920.969, 'duration': 9.024}, {'end': 7942.361, 'text': 'both in their ability to have compassion for other humans and their ability to maximize love in the world between humans.', 'start': 7929.993, 'duration': 12.368}], 'summary': 'Ai should be engineered for compassion and love.', 'duration': 29.488, 'max_score': 7912.873, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/Z1KwkpTUbkg/pics/Z1KwkpTUbkg7912873.jpg'}, {'end': 8122.836, 'src': 'embed', 'start': 8099.24, 'weight': 4, 'content': [{'end': 8109.507, 'text': 'um. so, from that perspective, um, i would hope that our work not just me, but our work in general um, people would look back and said hey,', 'start': 8099.24, 'duration': 10.267}, {'end': 8111.889, 'text': 'they really helped make this better future happen sooner.', 'start': 8109.507, 'duration': 2.382}, {'end': 8118.953, 'text': 'They helped us understand the nature of false beliefs sooner than we might have.', 'start': 8114.61, 'duration': 4.343}, {'end': 8122.836, 'text': "Now we're so happy that we have these intelligent machines doing these things, helping us,", 'start': 8119.213, 'duration': 3.623}], 'summary': 'The work aims to accelerate a better future by understanding false beliefs and leveraging intelligent machines.', 'duration': 23.596, 'max_score': 8099.24, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/Z1KwkpTUbkg/pics/Z1KwkpTUbkg8099240.jpg'}], 'start': 7100.404, 'title': 'Principles of flight and computation, career advice, and love in human civilization', 'summary': 'Discusses how understanding principles of flight and computation leads to innovations, emphasizes career advice and finding passion, and explores the role of love and compassion in human civilization, with examples of commercial applications and the impact on ai systems.', 'chapters': [{'end': 7365.585, 'start': 7100.404, 'title': 'Principles of flight and computation', 'summary': 'Discusses how understanding the principles of flight and computation can lead to innovations in the field, including commercial applications, such as speeding up deep learning networks by a factor of 100 and enabling continuous learning in neural networks.', 'duration': 265.181, 'highlights': ['Understanding the principles of flight and computation allows for innovative implementations, potentially leading to commercial applications like speeding up deep learning networks by a factor of 100.', 'Applying biological principles to machine learning techniques, such as sparsity and dendritic models, can significantly improve existing systems commercially.', 'Continuous learning and rapid learning can be achieved by adding dendrites to existing neural networks, potentially bringing the field of machine learning commercially closer to the goal of a true AI system.', 'Small innovations in neural networks, such as making them more asynchronous, dynamic, robust, and incentivizing sparsity, can lead to significant advancements, with potential for making things 100 times faster and reducing training costs for networks like transformer networks.', 'The effectiveness of deep learning for specific tasks is driving a lot of innovation in the hardware, which may lead to the discovery of intelligent systems operating differently or on a much larger scale than deep learning.']}, {'end': 7834.59, 'start': 7367.079, 'title': 'Career advice and finding passion', 'summary': "Delves into career advice and finding passion, emphasizing the importance of being passionate about one's work and the ability to overcome obstacles, illustrated by the speaker's journey in pursuing neuroscience and the parallels drawn with finding joy in various aspects of life.", 'duration': 467.511, 'highlights': ["The speaker's journey in pursuing neuroscience and the importance of passion and perseverance in overcoming obstacles, as exemplified by the challenges faced as a graduate student at Berkeley and in the computing industry. Importance of passion and perseverance in pursuing neuroscience, overcoming obstacles as a graduate student and in the computing industry.", 'Emphasizing the significance of finding joy and passion in various aspects of life, including pursuing intellectual endeavors, parenthood, and the mastery of a particular sport. Significance of finding joy and passion in intellectual endeavors, parenthood, and sports.', "The speaker's realization about the value of parenthood and the meaningful joy it brings, prompting a shift in perspective about the pursuit of passion and the allocation of time. Value of parenthood and the meaningful joy it brings, shift in perspective about the pursuit of passion and time allocation."]}, {'end': 8299.874, 'start': 7836.531, 'title': 'The role of love in human civilization', 'summary': 'Explores the significance of love and compassion in human civilization, emphasizing the need to engineer these qualities into ai systems for maximizing love between humans and the potential positive impact on human civilization.', 'duration': 463.343, 'highlights': ['The significance of love and compassion in human civilization is emphasized, with a focus on engineering these qualities into AI systems for maximizing love between humans. Emphasis on the importance of love and compassion in human civilization; Advocacy for engineering these qualities into AI systems for maximizing love between humans.', 'The potential positive impact of engineering love and compassion into AI systems is discussed, with a focus on the potential benefits for human civilization, such as solving the climate change problem and accelerating the arrival of a better future. Discussion on the potential benefits for human civilization, such as solving the climate change problem and accelerating the arrival of a better future.', "The speaker's hope that their work contributes to accelerating a better future for human civilization and understanding the nature of false beliefs, with the aim of making positive changes happen sooner. Desire for work to accelerate a better future for human civilization and understanding the nature of false beliefs; Aim to make positive changes happen sooner."]}], 'duration': 1199.47, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/Z1KwkpTUbkg/pics/Z1KwkpTUbkg7100404.jpg', 'highlights': ['Small innovations in neural networks can lead to significant advancements, making things 100 times faster and reducing training costs.', 'Applying biological principles to machine learning techniques can significantly improve existing systems commercially.', 'Understanding the principles of flight and computation allows for innovative implementations, potentially leading to commercial applications.', 'Emphasis on the importance of love and compassion in human civilization; Advocacy for engineering these qualities into AI systems for maximizing love between humans.', "The speaker's hope that their work contributes to accelerating a better future for human civilization and understanding the nature of false beliefs."]}], 'highlights': ['The neocortex, comprising 75% of the brain, continuously updates a model that includes memories of individuals and objects.', "The brain's ability to create, update, and share models of the world underlies the concept of collective intelligence, with language being a key part of this capability.", 'The neocortex accounts for 70-75% of the volume of a human brain, dominating in size and playing a crucial role in high-level functions like vision, language, planning, and thinking.', 'Most predictions occur inside individual neurons, especially the pyramidal cells, through internal dendritic spikes, providing insight into the neural mechanism of prediction.', 'The impact of AI on society is predicted to be as large, if not larger, than computing in the last century, due to its fundamental learning principle, creating a need to consider the integration of intelligent systems into the human world and the concept of consciousness in machines, challenging the notion of assigning desires and goals to intelligent machines.', 'The challenges of uploading human minds into digital space and the impracticality of replicating the entire body are discussed.', 'Small innovations in neural networks can lead to significant advancements, making things 100 times faster and reducing training costs.', 'Emphasis on the importance of love and compassion in human civilization; Advocacy for engineering these qualities into AI systems for maximizing love between humans.']}