Loading transcript...
Click for full transcript
Chapters & Sections (167)
▼
00:00
Host Greeting and Podcast Setup
01:05
Introduction of Jeffrey Hinton as AI Pioneer
02:09
Hinton’s Nobel Prize and Early Work
02:42
From Keyword Search to Contextual Understanding
03:44
Limitations of Traditional Search Engines
04:15
Large Language Models as Emerging Experts
04:50
Distinguishing Machine Learning from Neural Networks
05:51
Neural Network Foundations and Brain Connection
06:52
Neuron Activation Mechanics
07:55
Cooperative Neuron Coalitions and Decision Making
08:26
Cooperative Neuron Ping in Object Recognition
09:29
Overlap of Neural Coalitions for Similar Objects
10:00
General vs Specific Neuronal Firing Patterns
11:02
Layered Activation in Sensory and Language Areas
12:05
Programming Neural Nets for Adaptive Learning
15:42
Hebbian Rule and Synaptic Strength Adjustment
16:13
Stabilizing Neural Networks with Decay Mechanisms
16:45
Building a Multi‑Layer Network for Image Classification
17:16
Pixel Intensity as Neural Input
17:49
Limitations of Raw Pixel Information
18:19
Motivation for Edge Detection
18:50
Designing a Vertical Edge Neuron
19:20
Teaching the Network to See
19:50
Layered Edge Detectors and Weighting
20:22
Positive and Negative Connections for Contrast
21:23
Scaling Up Edge Detectors Across the Image
21:55
Combining Edges into Simple Shapes
22:28
Detecting Point‑Like Features (e.g., Beaks)
24:03
Extending Vision to Other Senses
25:34
Edge and Shape Detection in Early Layers
26:04
Combining Features to Recognize Bird Heads
27:06
Challenges of Manual Wiring and the Need for Automation
27:37
Random Initialization and Baseline Predictions
29:08
Iterative Manual Adjustment of Connection Strengths
30:09
The Computational Explosion of Manual Tuning
31:12
Backpropagation as a Unified Optimization Tool
31:43
Historical Context and the 1986 Breakthrough
33:16
Scaling Up Data and Computation for Real-World Vision
34:19
Factor of a Million Reduction in Transistor Area
35:23
From Desktop to Pen-Size Calculators
36:25
Training a Bird Detection Network
36:57
Emergence of Edge and Feature Detectors
38:01
Data and Compute Constraints in Early AI
39:03
Encoding Words as Neural Activations
39:33
Predicting the Next Word in Context
41:11
Backpropagation of Prediction Errors
42:46
Feature Activation Patterns in Language Models
43:17
Predicting the Next Word and Meaning Construction
43:48
Emotional and Moral Filtering in Neural Nets
44:19
Rapid Automatic vs Deliberate Processing
46:24
Distributed Data Exposure and Averaging Updates
48:28
Iterative Generation and Dopamine Feedback
49:29
Human Reinforcement Learning in Practice
52:06
Model Personality Switching for Word Prediction
52:37
Human Control as the Primary Threat to AI
53:08
Example of AI Hesitation and Human Interpretation
53:39
Distinguishing Risks from AI Itself vs. Its Use
54:11
Misuse of AI for Election Corruption
54:42
Brexit and Cambridge Analytica Data Exploitation
55:14
AI as a Modern Persuasion Tool
56:06
Model Agnosticism and Human Reinforcement
56:46
AI's Childlike Desire for Approval
57:17
Weaponized AI in Elections and Nerve Agent Development
57:48
Historical U.S. Election Interference Techniques
1:00:56
Experts Predict AI Will Surpass Human Intelligence in Two Decades
1:01:29
Potential Dangers of a Super‑Intelligent AI
1:02:00
Uncertainty and Positive Outlooks for AI Applications
1:02:33
Economic Disruption and Workforce Replacement
1:03:04
Monopolization Concerns and Power Dynamics
1:03:35
Balancing Benefits and Risks of AI Development
1:04:08
Money, Power, and Regulatory Challenges
1:05:10
Need for Democratic Governance in AI Regulation
1:05:41
China’s Approach to AI Governance and Control
1:06:45
European Efforts and Limitations in AI Regulation
1:07:46
International Collaboration to Prevent Existential Threats
1:08:17
China’s Strategic Use of AI for Global Dominance
1:08:49
Potential for Cooperative AI Safety Measures
1:09:52
AI as a Strategic Weapon and Collaboration Dynamics
1:10:22
Balancing AI Development with Human Oversight
1:10:53
China, Europe, and Global AI Leadership
1:11:23
Impact of Cutting Basic Science Funding on National Strength
1:11:53
Sustained Basic Research as the Foundation of AI Breakthroughs
1:12:25
Comparing U.S. and Chinese AI Funding Strategies
1:12:58
Startup Ecosystems and Government Support in China
1:13:29
Industry Perceptions of AI Risks and Opportunities
1:14:32
Ethical Concerns Around Public Release of AI Models
1:15:34
Monetization and Commercial Viability of AI Platforms
1:16:04
Weaponization and Sentient AI Risks
1:17:06
Defining Sentience and Its Implications for AI Behavior
1:18:09
Scientific Evidence of Self-Awareness in AI Systems
1:18:40
Awareness vs Consciousness
1:19:12
Flat Earth Analogy for Mental Models
1:19:46
Hallucination Example with Pink Elephants
1:21:00
Rejecting the Theater Metaphor
1:21:48
Perceptual System Misreporting
1:22:19
Subjective Experience as Malfunction Signal
1:23:30
Chatbot Demonstration with Prism
1:24:23
AI's False Self-Beliefs
1:25:58
Digital Immortality and Religious Implications
1:27:03
Defining Genuine Resurrection
1:27:36
AI Persuasion Capabilities
1:28:07
Metaphorical Perspectives on AI
1:28:39
Historical Context of Persuasion and AI
1:29:11
Emerging Technological Threats
1:29:42
Energy Consumption and Economic Impact
1:30:13
Risk Assessment of AI Threats
1:30:43
Acknowledging Expert Contributions
1:31:14
Historical AI Development Insights
1:32:17
Regulatory Challenges and Corporate Influence
1:33:18
Comparisons to Nuclear and Biological Threats
1:33:49
Public Awareness and Tipping Points
1:34:19
Cultural References to AI and Climate Change
1:34:50
Political Dynamics and Public Sentiment
1:35:52
Pardon Hope for High-Profile Figures
1:36:07
Critique of Political Rhetoric
1:36:25
Deflection in Media Interviews
1:36:45
Questioning Allegations and Responses
1:37:28
Social Media Promotion
1:37:45
Production Credits and Thank Yous
1h 37m⏭ 3
1h 22m⏭ 3
1h 28m⏭ 4
1h 33m⏭ 5
1h 35m⏭ 3
1h 12m⏭ 5
48m⏭ 1
1h 20m
3h 13m⏭ 2
1h 40m⏭ 3
31m⏭ 1
4h 6m⏭ 2
1h 22m
1h 16m⏭ 4
1h 50m⏭ 3
1h 0m⏭ 1
47m
18m