Understanding What AI Cannot Do

The Limits of Artificial Intelligence, Irreplaceable Human Capabilities and Strategies for Thriving in the Future of Work

Introduction

Chapter 1: The Promise and the Illusion - What Modern AI Actually Is

1. Where AI Stuns and Where It Stumbles: Real-World Examples
2. Intelligence, Cognition and What the Word Has Come to Mean
3. The Hype Cycle: Why Every Breakthrough Feels Permanent
4. What This Book Will Show You: A Roadmap of Limits and Opportunities

Chapter 2: Inside the Machine - How Contemporary AI Systems Work

1. Neural Networks: From Perceptrons to Deep Learning
2. Large Language Models and the Transformer Architecture
3. Generative AI: Image, Audio, Video and Beyond
4. Reinforcement Learning and Autonomous Agents
5. The Data Dependency: Training, Fine-Tuning and the Limits of Scaling

Chapter 3: Pattern Matching Without Comprehension

1. Next-Token Prediction: The Surprising Power of a Simple Task
2. When Pattern Matching Works: Breadth, Speed and Familiarity
3. The Hollow Center: Why Syntax Without Semantics Fails
4. Experiments That Reveal the Lack of Understanding
5. Can Scaling Lead to Emergent Understanding, or Just Better Mimicry?

Chapter 4: Reasoning Under Fire - Logical Consistency and Its Limits

1. The Rise of Reasoning Models: From Chatbots to Chain-of-Thought
2. When AI Gets Logic Right: Performance on Standardized Tests
3. Silent Errors: How AI Fails at Reasoning Without Showing It
4. The Problem of Consistency: Same Question, Contradictory Answers
5. Where Reasoning Limits Matter Most: Law, Science and High-Stakes Decisions

Chapter 5: Common Sense, Causality and the Missing World Model

1. What Common Sense Actually Is: The Hidden Knowledge of Everyday Life
2. AI Fails at Things Humans Never Learn: Physical Intuition and Social Norms
3. Correlation Without Causation: Why AI Confuses Association and Mechanism
4. The Winograd Schema and Other Tests of Implicit Knowledge
5. Can Causal AI Close This Gap, or Is It a Deeper Problem?

Chapter 6: Originality, Creativity and the Question of New Ideas

1. Defining Creativity: Recombination, Insight and Genuine Novelty
2. How AI Creates: Training Data, Latent Space and Stochastic Sampling
3. AI-Generated Art, Music, Code and Writing: Impressive or Derivative?
4. The Source Problem: Training on Human Work, Competing With Humans
5. Creativity in Teams: When AI Inspires and When It Flattens

Chapter 7: Emotional Intelligence, Trust and the Human Touch

1. What Emotional Intelligence Actually Involves: Perception, Response and Authenticity
2. AI and Emotional Detection: Technology That Reads Faces, Text and Voice
3. The Trust Deficit: Why People Want Humans Behind Important Decisions
4. Relationships, Empathy and the Problem of Intentionality
5. Care Work, Therapy and the Irreplaceability of Human Connection

Chapter 8: Moral Judgment, Ethics and the Problem of Responsibility

1. Moral Reasoning: Why Ethics Is Harder Than Rule-Following
2. How AI Handles Ethical Dilemmas: Training, Alignment and Guardrails
3. The Accountability Gap: Who Is Responsible When AI Fails?
4. Legal Liability, Professional Duty and the Limits of Automation
5. Bias, Fairness and the Illusion of Neutral Algorithms

Chapter 9: The Physical World - Dexterity, Perception and Embodiment

1. The Hard Problem of Robotics: Why Moving Objects Is Surprisingly Difficult
2. Breakthroughs in Robot Learning: From Controlled Labs to Real Environments
3. The Skilled Trades Advantage: Craftsmanship, Adaptability and Judgment
4. Sensory Integration: How Humans Perceive and Why AI Struggles
5. Economics of Physical Automation: When Robots Make Sense and When They Do Not

Chapter 10: Reliability in the Real World - When AI Fails and Why It Matters

1. Categories of AI Failure: Hallucination, Drift, Manipulation and Overconfidence
2. High-Stakes Mistakes: Healthcare, Law, Finance and Public Safety
3. Adversarial Vulnerabilities: How Small Changes Break AI Systems
4. The Cost of Errors: Quantifying the Price of AI Mistakes
5. Oversight and Guardrails: What Humans Must Monitor and When

Chapter 11: Augmentation vs Replacement - Mapping the Human-AI Boundary

1. A Typology of Work: Automatable, Augmentable and Human-Only Tasks
2. When Augmentation Beats Automation: Multiplying Human Capability
3. The Automation Threshold: Technical Feasibility, Economic Sense and Social Acceptance
4. Measuring the Complementarity: Evidence From Early Adopters
5. Risks of Over-Automation: When Delegating to AI Creates Vulnerability

Chapter 12: The Task vs The Job - Why Automation Does Not Equal Job Elimination

1. Jobs as Bundles of Tasks: Deconstructing Work
2. Historical Patterns: How Past Technologies Changed Jobs Rather Than Erasing Them
3. Task Displacement, Task Creation and the Reshaping of Roles
4. Why Some Jobs Shrink, Some Grow and Some Disappear
5. The Pace Factor: Why Transition Speed Matters Even If Jobs Evolve

Chapter 13: The Knowledge Economy Under Pressure - Professionals, Creatives and Analysts

1. Software Development: AI-Assisted Coding and the Changing Developer Role
2. Law and Legal Services: Document Review, Research and the Human Judgment Core
3. Finance and Consulting: Analysis, Advice and the Trust Premium
4. Research and Academia: AI in Science, Authorship and Knowledge Creation
5. Creative and Media Professions: Design, Writing and Content at Scale

Chapter 14: Service, Care and Skilled Work - Where Humans Remain Central

1. Healthcare: Diagnosis, Treatment Decisions and the Patient Relationship
2. Education and Training: Personalization, Mentorship and Human Motivation
3. The Skilled Trades: Construction, Repair and On-Site Adaptability
4. Hospitality, Retail and Customer Experience: Efficiency vs Authenticity
5. Public Safety, Government and Institutional Roles

Chapter 15: Vulnerability and Resilience - Assessing Where You Stand

1. Eight Criteria for Automation Risk: Task Composition, Predictability and More
2. Mapping Your Role: A Self-Assessment Approach
3. Industry-Level Vulnerability: Which Sectors Face the Most Pressure
4. Geographic and Institutional Factors: Why Risk Varies by Location
5. Resilience Signals: What Makes a Role or Organization Harder to Automate

Chapter 16: The Skills That Will Matter - Building a Durable Advantage

1. Critical Evaluation: Judging AI Output and Independent Thinking
2. Domain Expertise: Deep Knowledge as an AI Multiplier
3. Complex Problem-Solving: Defining Problems AI Cannot Define for You
4. Communication, Negotiation and Influence: High-Stakes Human Interaction
5. Interdisciplinary Synthesis and Systems Thinking
6. AI Literacy and Technical Competence: Using Tools Without Being Used By Them

Chapter 17: Career Resilience in Practice - Strategies Across Stages

1. For Students: Choosing Education That Future-Proofs You
2. Early Career: Building Foundations That Matter
3. Mid-Career Adaptation: Pivoting Without Starting Over
4. For Managers and Executives: Redesigning Teams and Workflows
5. Career Changers and Entrepreneurs: Finding New Paths in the AI Economy

Chapter 18: The Economic Landscape - Productivity, Inequality and Power

1. Productivity Promises: Why Gains Are Slower and Uneven Than Expected
2. Job Displacement and Creation: What the Evidence Actually Shows
3. Wage Inequality and Labor Market Polarization
4. Corporate Concentration: Who Captures the Value AI Creates
5. Global Inequality: Advanced Economies, Developing Nations and the Digital Divide

Chapter 19: Lessons From History - What Past Transitions Teach and Fail to Teach

1. The Industrial Revolution: Job Destruction, Job Creation and Decades of Pain
2. Electrification and Automation: Task Transformation in Manufacturing
3. The Computing and Internet Revolutions: New Economies, New Vulnerabilities
4. Patterns That Hold: Creativity, Adaptation and the Resilience of Labor
5. Why AI Might Be Different: Speed, Generality and Cognitive Labor

Chapter 20: Competing Forecasts - What Experts Predict and Why They Disagree

1. The Spectrum of Predictions: From Mild Disruption to Massive Displacement
2. Key Studies and Their Methodologies: How Projections Are Built
3. Areas of Expert Consensus: What Most Analysts Agree On
4. Points of Deep Disagreement: Timing, Scope and Irreversibility
5. Hidden Assumptions: Economics, Regulation and Human Adaptation

Chapter 21: Scenarios for the Next Two Decades - Preparing for Multiple Futures

1. Scenario One: Gradual Integration and Human-AI Collaboration
2. Scenario Two: Accelerated Automation and Sectoral Disruption
3. Scenario Three: Platform Economies and the AI-Augmented Entrepreneur
4. Scenario Four: Regulatory Slowdown and Managed Transition
5. Leading Indicators: What to Watch as Scenarios Unfold

Chapter 22: Your Strategic Roadmap - Principles for Thriving Alongside AI

1. Principles That Hold Across All Futures
2. Priorities for the Next Twelve Months
3. The One-to-Five-Year Horizon: Building Depth and Breadth
4. Five to Ten Years and Beyond: Adaptability as a Core Strategy
5. A Final Word: Agency, Responsibility and the Human Future

Conclusion

References