The New Competitive Landscape
Every business strategy book you've read is partially obsolete.
Not because the principles are wrong—differentiation, cost leadership, network effects—but because the context has changed. When AI can write code, generate designs, analyze markets, and interact with customers, the barriers that protected competitive positions erode. When global talent can be orchestrated remotely and AI can amplify small teams, scale advantages diminish. When new competitors can emerge in months rather than years, strategic planning horizons compress.
This chapter is for business leaders navigating transformation. Not a complete strategy manual—those fill libraries—but a framework for thinking about competition, operations, and survival when the rules are rewriting themselves.
The honest answer: nobody knows exactly what works. But some principles seem robust enough to warrant attention.
The Moat Problem
Traditional Moats
Brand: Built over decades. Recognition, trust, associations.
Scale: Cost advantages from volume. Manufacturing, distribution, purchasing power.
Network effects: Value increases with users. Platforms, marketplaces, social networks.
Switching costs: Hard for customers to leave. Data lock-in, learning curves, integration.
Intellectual property: Patents, trade secrets, proprietary technology.
Regulatory capture: Licenses, approvals, relationships that limit competition.
How AI Erodes Moats
Brand: AI-generated content floods information space. Trust becomes harder to establish. New entrants can simulate established presence.
Scale: AI enables small teams to achieve what required large organizations. Manufacturing moves toward distributed, flexible production.
Network effects: AI can bootstrap cold-start problems. Synthetic data, simulated users, cross-platform portability.
Switching costs: AI makes migration easier. Data portability, rapid integration, learning curves flattened.
IP: AI accelerates reverse engineering. Trade secrets harder to keep. Patents face validity challenges.
Regulatory: Still durable—but pressure for reform. New entrants lobby for access.
What Still Works
Data moats: Unique, proprietary data that AI can learn from. Customer behavior, operational insights, specialized domains.
Trust moats: Authentic relationships. Verified quality. Track record that can't be synthesized.
Speed moats: Faster iteration cycles. Learning faster than competitors.
Talent moats: Best people want to work for you. Culture that retains.
Integration moats: Deep embedding in customer operations. Hard to extract.
Regulatory moats: Still work—until they don't. Can buy time, not permanence.
AI-Native Operations
What "AI-Native" Means
Not: Adding AI to existing processes. Automating what you already do.
Rather: Redesigning processes around AI capabilities. Rethinking what's possible.
Example: Traditional customer service adds chatbot. AI-native customer service designs entire experience assuming AI handles 90%+, humans handle exceptions.
Operational Principles
Experiment velocity: How fast can you test ideas? AI enables rapid prototyping, A/B testing at scale, fast iteration.
Data as asset: Every operation generates data. Capture it. Clean it. Use it to improve.
Human-AI collaboration: Design workflows for augmented humans, not humans replaced by AI or AI working alone.
Continuous learning: Systems that improve from experience. Feedback loops that actually close.
What Changes
Product development: From long cycles to continuous iteration. From big bets to many small experiments.
Marketing: From campaigns to personalization at scale. From creative intuition to tested variations.
Operations: From standardized processes to adaptive systems. From manual monitoring to AI-detected anomalies.
Customer service: From call centers to AI-first with human escalation.
Finance: From periodic reporting to real-time visibility. From historical analysis to predictive modeling.
HR: From manual screening to AI-assisted matching. From annual reviews to continuous feedback.
Strategic Planning Under Uncertainty
The Problem
Traditional planning: 3-5 year strategic plans. Detailed forecasts. Clear milestones.
Current reality: Technology changes faster than planning cycles. Forecasts are unreliable. Plans are obsolete when completed.
Alternative Approaches
Scenario planning: Instead of single forecasts, multiple scenarios. Plan for range of futures.
Options thinking: Investments that create optionality. Small bets that could scale.
Adaptive strategy: Direction more than destination. Principles more than plans.
Fast experimentation: Test hypotheses quickly. Learn from reality.
Planning Horizons
Near-term (6-18 months): Can plan with reasonable confidence. Execute against clear objectives.
Medium-term (18-36 months): Direction and priorities. Flexible on specifics.
Long-term (36+ months): Scenarios and options. Prepare for multiple futures.
What to Plan For
Optimistic: AI delivers efficiency gains. You capture value. Competitors slower to adapt.
Baseline: AI transforms industry. Major adjustment required. Some winners, some losers.
Pessimistic: Disruption faster than adaptation. Current business model becomes unviable.
Wild: Complete industry restructuring. New competitors from unexpected directions.
Talent Strategy
The Talent Challenge
AI skills scarce: Everyone wants AI talent. Competition intense.
Skills evolve rapidly: What's valuable today may not be in three years.
Culture matters: Best people have options. Why work for you?
Principles
Buy vs. build: Some skills must be hired. Others can be developed internally.
Partner vs. own: Some capabilities better accessed through vendors, partners.
Generalists vs. specialists: Specialists for cutting edge. Generalists for adaptation.
Retention: Keeping good people harder than finding them.
What to Look For
Learning ability: Can they adapt as technology changes?
AI collaboration: Can they work effectively with AI tools?
Judgment: Can they evaluate AI outputs? Know when to trust?
Communication: Can they translate between technical and business?
Culture for AI Era
Psychological safety: People must be comfortable experimenting, failing, learning.
Continuous learning: Organization supports ongoing skill development.
Cross-functional: Boundaries between functions blur. Need people who can span.
Ethical awareness: AI raises questions. Culture that takes them seriously.
Technology Strategy
Build vs. Buy vs. Partner
Build: Custom solutions where you have competitive advantage. Where AI becomes core moat.
Buy: Commodity capabilities. Where vendors are better. Where speed matters.
Partner: Where you lack capability but need integration. Strategic relationships.
Platform Choices
Foundation models: Which providers? OpenAI, Anthropic, Google, open-source?
Cloud infrastructure: Where to run? AWS, Azure, GCP, hybrid?
Data platforms: How to manage data assets? Integration, governance, access.
Application layer: Build applications or use existing? Customization vs. speed.
Principles
Avoid lock-in: Where possible, maintain optionality. Standards, portability, multiple vendors.
Manage dependencies: Know what you depend on. Have contingencies.
Security first: AI creates new attack surfaces. Build security in from start.
Compliance by design: Regulatory requirements are coming. Easier to design in than retrofit.
Risk Management
AI-Specific Risks
Model failure: AI systems producing wrong outputs. Hallucinations, errors, biases.
Security: AI-powered attacks. Prompt injection. Data poisoning.
Regulatory: New laws. Liability. Compliance requirements.
Reputation: AI failures visible. Public reaction to AI decisions.
Dependency: Over-reliance on AI systems. Single points of failure.
Risk Management Approaches
Testing: Extensive testing of AI systems before deployment. Red-teaming. Adversarial testing.
Monitoring: Continuous monitoring of AI performance. Anomaly detection. Drift detection.
Human oversight: Humans in loop for high-stakes decisions. Escalation paths.
Audit trails: Document AI decisions. Enable review. Support compliance.
Incident response: Plans for when AI fails. Communication. Remediation.
Specific Considerations
Liability: Who is responsible for AI decisions? Clear accountability.
Insurance: Coverage for AI-related failures. Evolving market.
Contracts: AI terms with vendors. Data use. Performance guarantees.
Documentation: Records of AI systems, training data, and decisions are becoming legally required. The EU AI Act (effective 2025-2026) mandates technical documentation, risk assessments, and human oversight records for high-risk AI systems. The US Executive Order on AI (October 2023) establishes reporting requirements for frontier models. Sector-specific rules compound these obligations: financial services firms must document model risk management under SR 11-7, healthcare AI requires FDA clearance documentation, and hiring algorithms face audit requirements under laws like New York City's Local Law 144. Even where regulation hasn't yet caught up, maintaining thorough documentation of AI system design, training data provenance, performance metrics, and decision logs is essential for defending against future liability claims and demonstrating responsible use to customers, partners, and regulators.
Industry-Specific Considerations
Software and Technology
Opportunity: AI-enhanced products. Development productivity. New capabilities.
Threat: AI commoditizes basic software. Competition from AI-native startups.
Strategy: Move up value chain. Build on AI rather than compete with it.
Professional Services
Opportunity: AI-augmented professionals. Higher productivity. New offerings.
Threat: AI replaces junior work. Pyramid model stressed.
Strategy: Redesign service delivery. Focus on judgment, relationship, complexity.
Manufacturing
Opportunity: AI-optimized operations. Predictive maintenance. Quality control.
Threat: Smart factories favor new entrants. Distributed manufacturing.
Strategy: Invest in AI operations. Build data advantages. Flexibility.
Retail and Consumer
Opportunity: Personalization. Inventory optimization. Customer experience.
Threat: AI-native competitors. Reduced differentiation.
Strategy: Data moats from customer relationships. Trust and authenticity.
Financial Services
Opportunity: AI-enhanced analysis. Fraud detection. Personalized products.
Threat: Regulatory burden. Fintech competition. Disintermediation.
Strategy: Regulatory moat while it lasts. Data advantages. Trust.
Healthcare
Opportunity: AI-assisted diagnosis. Drug development. Administrative efficiency.
Threat: Slow adoption due to regulation. New entrants with AI-native offerings.
Strategy: Embrace AI for appropriate use cases. Build clinical evidence. Navigate regulation.
The Path Forward
Near-Term Actions (2026-2028)
Assess current state: Where are you on AI adoption? What's the gap?
Identify quick wins: Where can AI add value immediately? Low risk, clear benefit.
Build capability: Hire or develop AI skills. Partner where needed.
Experiment: Small pilots. Learn what works. Build organizational muscle.
Manage risk: Put governance in place. Security. Compliance. Oversight.
Medium-Term Positioning (2028-2032)
Scale what works: Expand successful pilots. Integrate into operations.
Redesign processes: Move from AI-added to AI-native. Rethink workflows.
Build moats: Invest in data, trust, speed advantages that AI enhances.
Watch for disruption: Monitor industry changes. Be ready to pivot.
Long-Term Preparation (2032+)
Scenario planning: What does your industry look like in 10 years? Multiple scenarios.
Options creation: Investments that create flexibility. Multiple bets.
Talent pipeline: Skills for the future. Continuous learning culture.
Existential questions: Does your business model survive? What needs to change?
Competitive Dynamics
Incumbent Advantages
Resources: Capital to invest. Existing cash flows.
Relationships: Customer relationships. Supplier relationships. Regulatory relationships.
Data: Historical data. Operational data. Customer data.
Talent: Existing workforce. Training infrastructure.
Incumbent Disadvantages
Legacy systems: Technical debt. Integration challenges.
Cultural inertia: "How we've always done it." Resistance to change.
Incentive misalignment: Short-term pressures. Quarterly earnings.
Cannibalization fear: New approaches threaten existing business.
Startup Advantages
Clean slate: Build AI-native from start. No legacy.
Speed: Faster decisions. Less bureaucracy.
Focus: Single mission. All-in on opportunity.
Talent: Attract people wanting to build something new.
Startup Disadvantages
Resources: Limited capital. Burn rate pressure.
Credibility: Unproven. No track record.
Distribution: Must build customer relationships from scratch.
Survivability: High failure rate. May not last long enough.
Competitive Implications
Fast-moving incumbents can win: Those who overcome inertia have advantages.
Focused startups can win: Those who solve real problems for customers.
Partnership often wins: Combinations of incumbent resources and startup innovation.
Confused middle loses: Neither incumbent advantages nor startup speed.
Governance and Ethics
Why It Matters
Regulatory pressure: Requirements coming. Better to be ahead.
Reputation: Public increasingly attentive to AI ethics.
Risk management: Ethical failures create operational risks.
Talent: Good people want to work for ethical organizations.
Governance Structures
Board oversight: AI strategy at board level. Risk committee attention.
Executive responsibility: Clear ownership. CAIO or equivalent.
Ethics review: Process for evaluating AI applications. Red lines.
Audit and compliance: Regular review. Documentation. External verification.
Ethical Considerations
Fairness: AI systems don't discriminate. Bias detection and mitigation.
Transparency: Stakeholders understand AI use. Explainability where required.
Privacy: Data use respects privacy. Consent. Data minimization.
Safety: AI systems are safe. Testing. Monitoring. Failsafes.
Human oversight: Appropriate human involvement. Not full automation of everything.
The Deeper Questions
What Is Your Business For?
If AI can do what you do, faster and cheaper, why do you exist? The answer must be something beyond efficiency—some value you create that matters.
What Do You Owe Your Workers?
If AI replaces roles, what is your responsibility to people who built your company? The ethics of displacement aren't just PR—they're about what kind of organization you want to be.
How Do You Compete Ethically?
If AI enables manipulation, surveillance, exploitation—and competitors use these tools—how do you compete while maintaining integrity?
What Future Are You Creating?
Your business choices shape the world. The technologies you deploy, the norms you establish, the culture you create—these matter beyond quarterly results.
Conclusion
Business strategy in the AI era is not fundamentally different from business strategy in any era: understand your environment, develop distinctive capabilities, serve customers well, adapt as conditions change.
What's different is speed and scope. Changes that took decades now take years. Competitive advantages erode faster. New competitors emerge from unexpected directions. The margin for error is smaller; the premium on adaptation is higher.
The businesses that thrive will be those that embrace AI as core to operations—not an add-on but integral to how they work. Those that build moats AI can't easily erode—data, trust, speed, integration. Those that attract and retain talent who can navigate change. Those that manage risk without being paralyzed by it.
There's no formula. Every industry, every company, every situation is different. But the principles are clear: experiment constantly, adapt quickly, invest in capabilities that compound, build relationships that last, and never assume today's position is permanent.
The businesses that survive the next decade will be those that took the next decade seriously while it was still the future. That's now.
Endnotes — Chapter 62
- Competitive moats: concept popularized by Warren Buffett; traditional moats include brand, scale, network effects, switching costs, patents, regulatory advantages.
- AI-native operations: designing processes around AI capabilities rather than adding AI to existing processes; requires fundamental rethinking of workflows.
- Foundation models: large pre-trained models (GPT-4, Claude, Gemini, etc.) that provide base capabilities; build vs. buy vs. partner decision for most enterprises.
- Scenario planning: strategic planning technique developed at Shell; considers multiple possible futures rather than single forecast.
- AI governance: emerging corporate function; includes board oversight, ethics review, audit, compliance; various frameworks available (NIST, EU AI Act requirements).
- Model failure risks: AI systems can fail in various ways including hallucination, bias, brittleness to distribution shift, adversarial vulnerability.
- Compliance by design: building regulatory compliance into systems from beginning rather than retrofitting; reduces long-term costs and risks.
- Technical debt: accumulated cost of expedient decisions in software development; legacy systems often carry significant technical debt that impedes AI adoption.
- AI talent market: significant competition for AI specialists; 2024 salaries for senior ML engineers often $300-500K+; retention a major challenge.
- Ethical AI frameworks: various frameworks for responsible AI development including IEEE, Partnership on AI, Anthropic's Constitutional AI principles.