A Tale of Potential collaboration. Will Algorithms and Boardroom Decisions Meet

A Story of Strategic Evolution

In the heart of New York City, two consulting firms were tasked with the same challenge: design a market entry strategy for an American bagel brand in Paris. Firm A, a well-regarded strategy outfit, assembled a team of top-tier consultants who spent a week meticulously researching consumer preferences, mapping competitive landscapes, and devising a go-to-market plan. Firm B, a smaller but tech-savvy agency, took a different approach-leveraging AI-powered strategy models to draft a proposal in under an hour. When the client reviewed both strategies, the results were eye-opening. While the human-led team brought rich contextual understanding and deep consumer insights, the AI-assisted team introduced unique, data-backed ideas that challenged conventional thinking. The outcome? AI's computational power created a sharper, more agile strategic direction. This story potentially brings to the fore a larger transformation: Can AI actively shape high-stakes strategic decisions? Can machines truly master the art of strategy?

The Expanding Role of AI in Strategy

The notion of AI encroaching upon human judgment in strategic planning seemed implausible not long ago. But landmark events have changed the discourse. In 2016, AlphaGo's Move 37 stunned the world, demonstrating AI's ability to act strategically. More recently, an experiment comparing MBA students with an AI-driven strategy generator for a market entry plan yielded comparable results, proving AI's potential in formulating business strategy.

Today's AI models-such as GPT-4, Gemini, and Claude-are advancing beyond routine decision-making into the realm of strategic planning. They excel at pattern recognition, signal detection, and trend forecasting at scales that no human strategist can match. This poses a compelling question: Is AI ready for showtime? How can AI be integrated into strategy formulation that traditionally relied on human cognition, intuition, and foresight?

AI as a Strategic Partner: Automation Meets Augmentation

Historically, automation focused on routine and codifiable tasks. However, with the rise of Large Language Models (LLMs), AI now has the capacity to analyse vast, unstructured datasets and recombine them to generate novel strategic insights.

Unlike tactical decision-making, strategic choices often involve incomplete data and uncertain environments. AI can uncover non-obvious opportunities, simulate outcomes, and provide structured analyses in high-stakes situations. This shift represents a paradigm where AI can be used for tasks that go beyond merely making activities efficient but can identify novel pathways that can improve the competitive advantage of firms.

The declining costs of computational power and the surge in AI investments reinforce this trajectory. The Artificial Intelligence Index Report 2024 notes that AI-related projects on GitHub skyrocketed from 845 in 2011 to 1.8 million in 2023. As training costs for advanced models like GPT-4 ($78 million) and Gemini Ultra ($191 million) rise, so does the influx of capital into generative AI, reaching $25.2 billion in 2023 alone.

This exponential growth suggests that future AI models will possess even more sophisticated reasoning abilities. However, can they address the ultimate goal of strategy: how does one generate feasible yet novel ideas that provide a competitive advantage?

Human + AI: A Powerful Synergy

Strategic decisions require navigating complex domains where cause-and-effect relationships are often unclear. Business leaders frequently make high-risk choices without definitive answers, relying on intuition, experience, and judgment. Can AI discern a good strategy from a bad one?

Initial results suggest that it can. When GPT was tasked with evaluating startup business plans graded by venture capitalists, it outperformed entrepreneurs in assessing and refining ideas. However, AI's effectiveness may be limited for several reasons. In ambiguous domains, AI performs best only when guided by structured rubrics. While it can process vast amounts of data and reveal correlations, it lacks the lived experience and nuanced judgment necessary for high-stakes strategic decisions.

Also, when leveraging an LLM's capabilities, the aggregation of multiple imperfect predictions can lead to more accurate outcomes, as individual errors may cancel each other out. Recent research suggests this effect is particularly prominent in strategic decision-making, where predictions often contain some degree of inaccuracy.

Moreover, AI-driven aggregation often produces an insight that is, at best a weighted average of a set of strategic responses for a given problem. This may not always be useful since, to acquire and sustain a competitive advantage, firms need to identify and pursue a unique strategy that is non-replicable.

Finally, machine learning models can inherit biases from training data, resulting in flawed recommendations. A well-documented example was Amazon's AI-powered hiring tool, which displayed gender bias due to skewed historical data. In addition, AI's propensity for 'hallucinations'-where it generates plausible but incorrect insights-can pose risks if left unchecked.

The key challenge for managers, then, is to ensure that AI does not simply deliver statistical generalizations but actionable insights that are novel yet feasible. This is where human intuition and contextual expertise become essential. Managers must ask the right questions to extract meaningful, high-value insights rather than generic conclusions. Ultimately, while AI can enhance decision-making, human judgment remains crucial in selecting the most relevant and impactful course of action for an organization. This underscores why AI should be viewed as an enabler rather than a replacement for human decision-makers.

With AI support, managers can enhance their strategic capabilities-anticipating shifts, identifying hidden risks, and optimizing decisions. AI's capacity to analyze complex datasets without cognitive biases helps mitigate human tendencies like success bias, where decision-makers disproportionately focus on past successes (e.g., iPhone, Netflix) while neglecting failures (e.g., Google Glass, Windows Phone).

For human-AI collaboration in strategic decision-making to succeed, strategists must formulate the right questions, critically interpret AI-driven insights, and employ contextual judgment to enhance AI-generated solutions. The most effective decision-making model is not AI in isolation but a collaborative approach, where AI offers analytical depth while humans contribute wisdom and ethical considerations.

The case of AlphaGo's Move 37 exemplifies this. Initially seen as an erratic play, it was later recognized as a groundbreaking strategy, prompting Go players to adopt more dynamic styles. Similarly, AI's ability to challenge established business paradigms can encourage leaders to rethink conventional wisdom and explore innovative approaches.

This evolution is not about humans versus AI; it's about leveraging AI as a force multiplier. By integrating AI into strategic decision-making, organizations can enhance their ability to anticipate disruptions, evaluate opportunities, and navigate uncertainties with improved precision.

The Future of Strategy: A Collaborative Path Forward

Strategic advantage stems from making high-impact decisions while managing risk. AI can model scenarios, highlight probabilities, and uncover overlooked patterns, but the final call still rests on human judgment. The lesson is clear-AI offers powerful analytical support, but human decision-makers must apply contextual intelligence, ethical considerations, and long-term vision to strategic choices.

Just as AlphaGo inspired new approaches to Go, AI will continue to reshape business strategy. The future will not be defined by AI replacing human strategists but by organizations that learn to harness the best of both worlds-melding AI's analytical prowess with human creativity, experience, and intuition.

The key to winning in the AI era? Not choosing between humans or machines, but fostering a partnership where each amplifies the other's strengths.