Theory 01
Dynamic Capabilities Framework
Origin: Teece, Peteraf & Leih (2016); Session 4 slides
Teece et al. (2016) define Dynamic Capabilities as the firm's ability to integrate, build, and reconfigure internal and external competences to address rapidly changing environments. It's the capacity to remain competitive when "inflection points" emerge.
Sensing
Identifying and assessing opportunities (and threats) in the environment. Scanning the horizon for shifts like AI.
Seizing
Mobilizing resources to address the opportunity. Making the strategic "bet" and investing in the new path.
Transforming
Continuous renewal and reconfiguration of assets. Changing organizational structures and culture (Teece et al., 2016).
Theory 02
Ordinary vs. Dynamic Capabilities
Origin: Teece, Peteraf & Leih (2016); Session 4 slides
Session 4 distinguishes ordinary capabilities from dynamic capabilities. Ordinary capabilities help the firm perform current activities efficiently. Dynamic capabilities help the firm adapt, innovate, and reconfigure resources when the environment changes.
| Dimension | Ordinary Capabilities | Dynamic Capabilities |
|---|---|---|
| Purpose | Technical efficiency in existing business functions. | Strategic fit and evolutionary fitness over the long run. |
| Logic | Operational, administrative, and governance routines. | Sensing, seizing, shaping, transforming, and asset orchestration. |
| Imitability | Often easier to benchmark and imitate. | Harder to imitate because they depend on managerial judgment, history, culture, and resource orchestration. |
| Exam implication | Explains how the firm performs today. | Explains how the firm survives technological discontinuity. |
Agility Is Not Always Good
Teece et al. warn against treating agility as a universal virtue. Agility is costly and can sacrifice efficiency. The managerial task is to calibrate how much agility is needed, based on whether the firm faces ordinary risk or deep uncertainty.
Theory 03
AI as an Organizing Capability
Origin: Stelmaszak, Joshi & Constantiou (forthcoming/2026); Session 4 slides
Stelmaszak et al. (2026) propose an ontological shift: AI is not just an "entity" or a "tool". It is an Organizing Capability that arises from the relations between human and algorithmic actors.
Properties of AI Capability (Stelmaszak et al., 2026)
- Connective: AI links people, data, and processes across traditional silos.
- Codependent: Neither humans nor algorithms can act effectively without the other in complex tasks.
- Emergent: The resulting organizational intelligence is greater than the sum of its parts.
Managerial Implication
The wrong question is "How do we adopt AI?" The stronger question is "How do we design the relations that make AI work?" Managers must structure how domain experts, data, algorithms, workflows, accountability, and feedback interact.
Theory 04
Strategy in an Era of Abundant Expertise
Origin: Yerramilli-Rao et al. (2025); Session 4 slides
Generative AI changes the economics of expertise. Expertise remains a source of competitive advantage, but the cost of accessing some forms of expertise falls when AI copilots, chatbots, and assistants become widely available. This creates a strategic dilemma: firms must decide which expertise remains scarce and which expertise can be augmented or commoditized.
Expertise as Advantage
Deep theoretical knowledge and practical know-how still matter, especially in regulated and complex domains.
Abundant Access
AI tools reduce the cost of accessing generalized knowledge, code, analysis, writing, and decision support.
Judgment Premium
As AI provides more answers, human value shifts toward framing problems, validating outputs, and deciding what matters.
Strategic Questions from the Slides
- Which problems can customers now solve with AI themselves?
- Which types of expertise must evolve for the firm to remain ahead of AI's capabilities?
- Which assets can the firm build or augment to stay competitive as AI advances?
- Where can AI create cost and time savings, and where should saved resources be redeployed?
Theory 05
Uncertainty vs. Risk
Origin: Teece, Peteraf & Leih (2016)
Teece et al. (2016) distinguish between Risk (known outcomes with probabilities) and Deep Uncertainty (unknown unknowns). Strong dynamic capabilities are essential for addressing the latter.
Risk
Can be managed with traditional tools (insurance, hedges). Probabilities are calibrated (Teece et al., 2016).
Deep Uncertainty
Ubiquitous in innovation economies. No clear probabilities. Requires agility and asset orchestration (Teece et al., 2016).
| Environment | Management Logic | Strategic Tool |
|---|---|---|
| Risk | Known outcomes with probabilities. Can often be hedged, insured, contracted, or modeled. | Risk management, contracts, financial hedges, buffers, compliance. |
| Deep uncertainty | Unknown unknowns. Future technologies, markets, and competitors cannot be assigned reliable probabilities. | Dynamic capabilities, scenario thinking, experimentation, real options, asset orchestration. |
| Inflection point | A shift that changes the basis of competition. | Sensing early, seizing decisively, transforming without destroying useful assets. |
Case Application
Novo Nordisk — Dynamic Agility
Applying Session 04 concepts to the synopsis:
- Sensing the AI Inflection: Novo's partnership with OpenAI is a clear example of Sensing the disruptive potential of LLMs in biotech.
- Asset Orchestration: By integrating OpenAI's algorithms with their own bio-data, Novo is Seizing the opportunity and reconfiguring its R&D competence.
- Agility vs. Efficiency: The 9,000 layoffs represent a painful Transformation — sacrificing traditional organizational efficiency (human-centric silos) to build the agility required for an AI-led future.
Exam Preparation
Likely Oral Exam Questions
-
Core What are the three pillars of Dynamic Capabilities according to Teece? ▶
- Sensing: Scanning for opportunities/threats.
- Seizing: Mobilizing resources to act.
- Transforming: Reconfiguring the organization to sustain the change.
-
Core What is the difference between ordinary and dynamic capabilities? ▶
- Ordinary capabilities support efficiency in current operations.
- Dynamic capabilities support adaptation through sensing, seizing, and transforming.
- The exam move is to explain that efficiency is not enough under deep uncertainty; firms need the capacity to reconfigure resources.
-
Core Why does Stelmaszak et al. (2026) argue that AI is an "Organizing Capability" rather than an entity? ▶
- Traditional views see AI as a "tool" or "autonomous agent".
- Stelmaszak argues that intelligence *emerges* from the relation. Without human data, feedback, and context, the algorithm is useless. Without the algorithm, the human cannot process the scale of data.
- Strategic implication: Competitive advantage is in the *quality of the relation*, not just buying the best tech.
-
Apply How does abundant expertise change strategy? ▶
- AI lowers the cost of accessing generalized expertise.
- Firms must identify which expertise is becoming commoditized and which expertise remains scarce.
- Competitive advantage shifts toward problem framing, proprietary data, validation, judgment, and the ability to combine human and algorithmic expertise.
-
Synthesis How does "deep uncertainty" impact the "make or buy" decision for AI? ▶
- In conditions of deep uncertainty, building internally (Make) is risky because the tech evolves so fast.
- Partnering (Buy/Partner) provides Agility. It allows a firm like Novo to access cutting-edge tech (OpenAI) without being locked into a soon-to-be-obsolete internal system.
- However, agility comes at a cost of potential dependency (Session 08/TCT).