The Paradox of Action: Leading In AI’s Slipstream

Q: Is ‘Wait and See’ the Most Dangerous Strategy of All?

The window for building AI readiness is narrowing. The companies that start developing these capabilities today may have 12 to 18 months to establish meaningful competitive advantage before AI becomes commoditized across their sector.

John Taylor
Apr 1
6 min read

There is an AI paradox facing every leader. Investing in AI can feel futile when today's cutting-edge technology becomes obsolete in 6 months. But "wait and see" might be the riskiest strategy of all. There is a 12 to 18 month window to to build AI capabilities and "preparing intensively while acting patiently" will create lasting competitive advantage.

Beyond the Hype Cycle: While competitors rush into reactive AI implementations, discover the three strategic capabilities that separate AI leaders from those leading expensive failures.

Great leaders are respected for their decisive action at key moments, whether in business, battle, or public life. Rarely are there plaudits for inaction.

In general, leaders are action-oriented; they have a bias to action. Leaders are expected to make fast and difficult decisions while projecting confidence, even if the balance of evidence towards a course of action is unclear.

This bias towards action faces a severe challenge in the emerging AI age. AI is rapidly disrupting ever larger parts of business, private, and public life. And the pace of AI change is not diminishing. When mobile phones emerged, there was the ‘iPhone moment’ that created a new category of experience and usage. There were many further advances, but these were relatively gradual, over many years. It was possible for non-experts to understand the key technology, and to build robust business cases for applications and businesses that could last years. What started as a consumer-led change quickly impacted enterprises and their ways of working. There was huge investment and disruption, but at a pace that could generally be managed.

In their personal lives, most leaders eventually discarded their Blackberrys and become avid smartphone users, with all of the understanding this conferred. Leaders became mobile-savvy and increasingly able to lead strategic and investment decisions with confidence.

AI is different. There are no parallels in the history of technology with the speed of change, or the breadth of AI and its impacts. The intensity of resources devoted to advancing AI tools and models is creating ever greater capabilities, leaving even experts amazed by the what can be achieved today. Leaders are firmly in the AI Age.

The Paradox of Action: Why Smart Leaders Are Unsettled by AI’s Speed

The speed of AI change has created a paradox that I call ‘the paradox of action’. Why invest in AI if it is going to move so quickly that whatever you build or implement will be obsolete in 6 months?

The incredible progress of mainstream Large Language Models (LLMs), such as ChatGPT, is perhaps the starkest example of change. ChatGPT 3.5 was very impressive when it arrived in November 2022 and acquired users at an extraordinary rate, but the latest models from Open AI, Anthropic and others make version 3.5 look rather limited and error-strewn. And better models will be with us soon. Surely this makes a wait and see approach look prudent?

In Boardrooms in every sector of the economy AI is being discussed, sometimes from a position of knowledge and expertise, but more often within the context of concern and trepidation. What are competitors doing? How do we respond to customers’ questions about our AI offer? Is AI going to erode our business? How to staff feel about their futures?

AI is on many companies’ risk registers, but mitigation measures are barely adequate. This is understandable given sheer pace of developments, the complexity of AI technologies, and the mixed messages in the media and from analysts about where to invest.

For leaders the journey to AI maturity echoes the early days of Equality, Diversity, and Inclusion (EDI). It was essential to develop fluency before designing complex change and EDI programmes. But at least with EDI there is some stability; carefully crafted multistakeholder initiatives don’t become obsolete in months.

Is ‘Wait and See’ the Most Dangerous Strategy of All?

‘Wait and see’ seems to be a credible strategy. “I’ll wait for the iPhone moment and then I can invest.”

A company that last year invested heavily in training and fine tuning an open source LLM could well find that a new LLM is better ‘out of the box’. The rise of ‘zero shot’ and ‘few shot’ learning in which models can be trained with little or no training date, means that the traditionally resource-intensive process of gathering and managing training data is largely removed. This greatly reduces the resources and time needed to move from idea to implementation, and can render months of data collection, labelling, and training largely wasted.

The leader who made headlines in 2023 with AI investments may find themselves wondering why they bothered. They may have followed the advice of external consultants, or internal AI enthusiasts, or simply acted from their own belief, but the business benefits are absent. With 90% of AI pilots failing (McKinsey, 2024), then why invest at all?

In 2021, McDonald's began to experiment with AI automated voice ordering at its drive-throughs. However, after significant investment and public fanfare, McDonald's quietly discontinued the programme because of inaccurate order taking and customer complaints. This exemplifies reactive action, with a rush to implement AI when it wasn’t ready, and indicates a lack of strategic readiness and planning.

Why AI Transformation is Never Just Plug-and-Play

Compounding the case for inaction is the complexity of embracing AI and the seismic changes it brings.

AI has the potential to change all aspects of a business, from how it manages its back office to how it engages with end customers. Managing change of this scope requires activity across a number of key areas, from organisational structure and culture, to technology and product development. There are rare cases when an AI tool or model can be ‘plugged in’ to replace an existing process; far more often this requires comprehensive planning and the engagement of multiple departments and specialisms.

So not only is AI moving very rapidly, it is complex and time consuming to implement.

Given these challenges, the temptation is to wait indefinitely. But this creates a different kind of risk.

The Window For Action Is Narrowing

While strategic patience is essential, the window for building AI readiness is narrowing. The companies that start developing these capabilities today may have 12 to 18 months to establish meaningful competitive advantage before AI becomes commoditized across their sector.

This isn't about rushing into implementation, but recognising that building the organizational muscle for AI decision-making takes time. The executives who begin developing AI fluency now will be positioned to act decisively when their moment arrives. Those who wait for perfect clarity may find themselves perpetually behind, always preparing but never ready.

The question isn't whether your competitors will embrace AI, but whether you will be ready when they do.

The Path Forward: Strategic Readiness, not Reactive Action

The paradox of action isn’t solved by choosing action or inaction, but rather by choosing strategic readiness over reactive responses.

There are three key capabilities that position leaders to succeed:

AI Fluency at the Leadership Level: Leaders need sufficient understanding of AI to distinguish between genuine opportunities and AI hype. There is no need to become a technical expert, but AI fluency enables the judgement to ask the right questions, and to recognise when AI vendors are overselling their capabilities.

Organizational Learning Architecture: Instead of just placing big bets on specific AI tools or products, invest in your organisation's ability to rapidly evaluate, pilot, and scale AI applications. Build the organisational processes and culture for rapid experimentation and quick learning.

Strategic Patience with Tactical Agility: The companies that thrive will often be those that resist the pressure to act immediately but instead build the capability to act decisively when the right moment arrives. Sometimes this requires the courage to appear slower than competitors in the short-term in order to be faster in the long-term.

The leaders who master this paradox may not be remembered as the first to implement AI, but rather as the first to implement it well and at scale. In an age where AI changes relentlessly, the sustainable competitive advantage will go to the wisest actors, not necessarily the fastest ones.