Paths, Capabilities and Consequences
Introduction
Chapter 1: Defining Super Intelligence
- What Super Intelligence Means
- Artificial Intelligence and What It Is Not
- Why Precision Matters
- Operationalizing the Concept
- Scope and Boundaries
- Summary
Chapter 2: The History of Machine Intelligence
- Ancient Dreams of Artificial Beings
- The Foundations of Computation
- The Birth of Artificial Intelligence Research
- The First Disappointments
- Expert Systems and the Second Boom
- New Directions and Quiet Progress
- Deep Learning and the Modern AI Boom
- The Emergence of Super Intelligence as a Research Topic
- Lessons from History
Chapter 3: The Components of Intelligence
- Perception and Representation
- Learning
- Reasoning and Problem-Solving
- Memory and Knowledge
- Pattern Recognition
- Prediction and Modeling
- Decision-Making
- Creativity
- Communication and Social Intelligence
- Adaptation and Generalization
- Metacognition and Self-Correction
- Autonomous Goal Pursuit
- Which Capabilities Are Necessary?
Chapter 4: Theories of Intelligence
- The g Factor and General Intelligence
- Computational Theories of Mind
- Connectionism and Neural Network Theories
- Information-Theoretic Theories
- Emergentist Theories
- Evolutionary Theories
- Predictive Processing and Active Inference
- What These Theories Tell Us About Super Intelligence
Chapter 5: Technological Foundations
- Computational Hardware
- Neural Network Architectures
- Data and Knowledge
- Software and Infrastructure
- The Convergence of Factors
Chapter 6: Approaches to Developing Super Intelligence
- Scaling
- Advanced Reasoning
- Learning and Training Methods
- Memory and Knowledge Systems
- Multimodal Systems
- Agentic Systems and Tool Use
- Autonomous Research and Automated Scientific Discovery
- Recursive Improvement and AI-Assisted AI Development
- Summary of Approaches
Chapter 7: Architectures and Capabilities
- Architectural Requirements
- Capability Profile
- The Relationship Between Architecture and Capability
- Limitations and Open Questions
Chapter 8: Efficiency
- Computational Efficiency
- Data Efficiency
- Energy Efficiency
- Reasoning Efficiency
- Memory Efficiency
- Training Efficiency
- Inference Efficiency
- Algorithmic Efficiency
- Economic Efficiency
- Time Efficiency
- Trade-offs and Optimization
Chapter 9: Strengths of Super Intelligence
- Superior Reasoning
- Accelerated Learning
- Scientific Discovery
- Engineering and Design
- Prediction and Forecasting
- Strategic Planning
- Medical Research and Healthcare
- Education and Knowledge Transfer
- Automation and Productivity
- Large-Scale Problem-Solving
- Summary of Strengths
Chapter 10: Weaknesses and Limitations
- Reliability Problems
- Lack of Grounding
- Brittleness
- Poor Generalization
- Hidden Failure Modes
- Objective Misalignment
- Adversarial Vulnerabilities
- Cybersecurity Risks
- Excessive Autonomy
- Dependence on Infrastructure
- Difficulty of Verification
- Uncertainty
- Capability Without Wisdom
- Summary of Weaknesses
Chapter 11: Applications and Benefits
- Science
- Medicine and Biotechnology
- Engineering and Technology
- Climate Research and Environmental Science
- Energy
- Agriculture and Food Systems
- Education
- Transportation and Logistics
- Manufacturing
- Space Exploration
- Business and Economics
- Public Administration
- Summary of Benefits
Chapter 12: Risks and Failure Modes
- Technical Risks
- Alignment Risks
- Misuse Risks
- Accident Risks
- Economic Disruption
- Geopolitical Competition
- Autonomous Replication
- Loss of Human Control
- Systemic Dependence
- Distinguishing Demonstrated from Theoretical Risks
- Summary of Risks
Chapter 13: AI Alignment and Control
- The Alignment Problem
- Objective Specification
- Reward Design
- Scalable Oversight
- Interpretability
- Robustness
- Corrigibility
- Monitoring and Evaluation
- Adversarial Testing
- Human Feedback
- Containment
- Unresolved Research Questions
- Summary of Alignment and Control
Chapter 14: Self-Improvement and Recursive Capability Improvement
- What Is Self-Improvement?
- AI-Assisted AI Development
- Automated AI Research
- Recursive Self-Improvement
- Intelligence Explosion
- Gradual Improvement
- Intelligence Amplification
- Comparative Analysis
- Summary of Self-Improvement
Chapter 15: Economic and Social Implications
- Employment and Labor
- Productivity and Economic Growth
- Wealth Creation and Distribution
- Inequality
- Education
- Social Structures and Institutions
- Corporate Power
- Summary of Economic and Social Implications
Chapter 16: Governance and Geopolitics
- The Governance Challenge
- Current Governance Landscape
- Regulation of AI Development
- Regulation of AI Deployment
- International Competition and Cooperation
- Governance of Super Intelligence
- Corporate Governance
- Summary of Governance and Geopolitics
Chapter 17: Philosophy and Ethics
- What Is Intelligence?
- Consciousness
- Agency and Autonomy
- Values and Morality
- Human Uniqueness
- Machine Personhood
- The Relationship Between Humans and Super Intelligence
- Summary of Philosophy and Ethics
Chapter 18: Long-Term Implications and Future Scenarios
- Scenario 1: Beneficial Super Intelligence
- Scenario 2: Uneven Benefits
- Scenario 3: Misaligned Super Intelligence
- Scenario 4: Misuse
- Scenario 5: Intelligence Explosion
- Scenario 6: Gradual Transformation
- Scenario 7: Stalled Development
- Comparing Scenarios
- Uncertainty and Humility
Chapter 19: Unresolved Questions
- Will Super Intelligence Emerge?
- How Fast Will It Emerge?
- Can We Solve Alignment?
- Can We Ensure Safety?
- What Are the Limits of Interpretability?
- How Will Super Intelligence Affect Employment?
- How Should Super Intelligence Be Governed?
- Can We Prevent Existential Risk?
- What Moral Status Do Advanced AI Systems Have?
- How Should We Prioritize Research?
- Summary of Unresolved Questions
Conclusion
- What We Know
- What We Do Not Know
- The Factors That Matter
- A Framework for Thinking About Super Intelligence
- What Readers Should Know
- Final Thoughts