AI'S TWO WAVES

November 14, 2025 | RACHIT GARG

Corporate profits are strong, markets are hitting all-time highs, yet an atmosphere of uncertainty emanates from boardrooms and economic forums. CEOs are expressing unprecedented caution about the future, despite presiding over healthy balance sheets. This paradox defines our current moment. At the heart of it lies a single, powerful force: Artificial Intelligence.

We are told this is the dawn of a new era, Industry 5.0, as they call it. Still, the fundamental question remains: Is the rise of AI, particularly generative AI, a true revolution on par with the Industrial Revolution, or is it merely the next step in a long line of technological evolutions? The answer will not only shape investment strategies and corporate planning but will redefine the very fabric of our economic and social lives. To navigate the path forward, we must first understand the nature of the shock we are experiencing, look to history as a guide, and define our collective responsibilities in steering this powerful force.

The Economic Compass: Decoding the AI Shock

To understand the turbulence, we need a model. The classic Cobb-Douglas production function provides a powerful lens when we expand it:

The model describes how an economy combines Factors of Production, namely Capital K and Labour L, to create output Y. We are using this model because it has helped us understand the economic impact of historical events like the feminist movement and the industrial revolutions, which resulted in significant shifts in the Factors of Production or their efficiencies.

This model also allows us to distinguish between two types of technological progress:

• Ak​ (Capital Augmentation): Technology that makes capital (machines, software, infrastructure) more intelligent and autonomous. This is the essence of automation.

• Al ​(Labour Augmentation): Technology that makes a human worker more productive. This is the essence of empowerment.

Using this understanding, we can see that the AI story is not one, but two distinct revolutions.

The First Wave (c. 2014-2024): The Capital Augmentation Shock (Ak​)

The AI of the last decade—machine learning, predictive analytics, and robotic process automation—was primarily a capital-augmenting force. It made systems smarter. Amazon's automated warehouses, Google's ad-bidding platforms, and high-frequency trading algorithms are prime examples of capital becoming more productive with less direct labour. The "Capital Augmentation" shock explains several current trends. It has fueled the growth of platform monopolies, a trend intensified by AI's powerful increasing returns to scale, where data feedback loops benefit dominant players. This shock also led to the hollowing out of mid-level jobs as routine administrative and analytical tasks are increasingly automated. The economy is still adjusting to the effects of this wave, including increased inequality.

The Second Wave (Present Day): The Generative AI Labour Augmentation Shock (Al)

The advent of generative AI marks a dramatic shift in character. Unlike its predecessor, which optimised systems in the background, generative AI is a direct cognitive partner for the knowledge worker. It is a quintessential labour-augmenting technology.

A developer using GitHub Copilot, a marketer brainstorming with ChatGPT, or an analyst summarising data with an AI tool is not being replaced; their productivity is being multiplied. This explains the current economic paradox: while the disruptive effects of the first wave are still causing uncertainty, the promise of the second wave is fueling market optimism and record profits. At the same time, the promised labour augmentation of the second wave is forcing the CEOs to ask: Is the labour-to-capital allocation optimal? Can we “Do More, With Less”?

This Al​ shock has profound implications:

• Productivity and Prices: As labour productivity rises, the effective supply of labour for cognitive tasks increases dramatically. Theoretically, this should lower the labour price per production unit, making sophisticated services cheaper and more accessible.

• The Capital Stress Test: A more productive workforce needs better tools. This surge in demand for the capital that powers AI—from GPUs to data centres—is already visible. The meteoric rise of NVIDIA's stock is not just a market trend but a direct macroeconomic signal of capital stress in response to a labour augmentation boom.

• The Wage Dilemma: For employees, the future is binary. If workers can leverage AI to increase their productivity at or above the pace of the industry, their value and salary should rise. Senior engineers who can now manage more complex systems are a prime example. Conversely, individuals in roles susceptible to direct AI replacement, such as junior coding, might experience wage stagnation or job displacement.

Historical Parallels and the Four-Phase Timeline

History does not repeat, but it rhymes. The AI revolution can be better understood by looking at past technological shocks, like the Industrial and Digital Revolutions. These events teach us two things: first, the real impact takes decades to manifest fully, and second, human ingenuity is remarkably adept at creating new wants and new industries to absorb productivity gains.

The Industrial Revolution, a capital-augmenting (Ak) shock, initially led to widespread labour displacement and soaring inequality. Yet, over time, it gave rise to entirely new sectors—from mass transportation to consumer goods—that created new roles for labour. Similarly, the internet revolution, a mix of Ak and Al​, didn't just automate old jobs; it created the digital economy, new forms of entertainment, and a greater focus on work-life balance, effectively inventing new ways to deploy our time.

We can model this process with a four-phase timeline for technological revolutions:

1. Installation & Hype: The technology emerges. Investment is speculative, and real-world use cases are limited.

2. Disruption & Adaptation: The technology begins to replace incumbent systems. This phase is marked by high uncertainty, job displacement, and a "productivity paradox" where gains are not yet visible in macro data.

3. Integration & Expansion: The technology becomes widespread. Society and business have adapted, and the productivity gains become clear and explosive. New industries are born.

4. Maturity & Saturation: The technology is fully integrated, and its growth rate slows to the background rate of the economy.

Where are we now? The first wave of AI (Ak​) is entering Phase 3, with its impact now clearly visible in the dominance of tech giants. Generative AI (Al​) is squarely in Phase 2: Disruption & Adaptation.

• The Next 5 Years: We can expect continued turbulence. The hollowing out of the mid-level job market will accelerate, affecting not just routine tasks but also cognitive roles in areas like law, marketing, and project management. Layoffs in these roles will coexist with soaring demand for AI-augmented talent. Inequality may worsen as the returns to capital and highly-skilled labour outpace the rest. The productivity gains will start to appear in specific sectors (like software development) but may not yet lift the entire economy.

• The Next 20 Years: As we move into Phase 3, society will have more fully adapted. New support industries, perhaps centred around sustainability, personalised healthcare, and live arts, will emerge to absorb the labour productivity surplus. Organisations will have to become more decentralised as AI empowers frontline forces. The nature of work will have fundamentally shifted from task execution to problem-finding, system design, and strategic oversight.

Navigating the Future: Responsibilities and Opportunities

The trajectory of the AI revolution is not predetermined. The choices we make today will shape our future. This calls for a clear definition of responsibilities for all stakeholders.

For Individuals: The Mandate of Lifelong Learning

The era of a single career based on a single degree is over. The primary individual responsibility is to cultivate adaptability and embrace continuous learning. The focus must shift from mastering specific tools to developing core skills that AI augments rather than replaces: critical thinking, creativity, emotional intelligence, and system-level design. The expectation for new hires will change; a college fresher in tech may now need to grasp system architecture from day one, as AI handles the basic coding.

For Organisations: Do More With More

The mantra of "Do More With Less" is a trap. AI offers an unprecedented spike in labour productivity, creating an excess of human capacity. The most forward-thinking organisations will ask, "How can we do more with this surplus?" Instead of cutting costs, they will redeploy this newly available talent toward innovation, deeper customer engagement, R&D, and tackling complex challenges like sustainability. This requires fundamentally redesigning job roles, moving from narrow task-based descriptions to broad, outcome-oriented responsibilities.

For Governments: The Agile Guide

Governments cannot command the direction of a revolution, but they can guide it. Their role is threefold:

1. Modernise Education: The curriculum must be radically updated to prepare students for an AI-augmented world. This means less rote memorisation and more focus on the "first principles" of problem-solving and creativity.

2. Create Robust Safety Nets: The transition will inevitably cause displacement. Strong social safety nets, including support for retraining, are essential. In the long term, if AI-driven automation significantly reduces overall employment, it could suppress consumer demand and lead to economic stagnation. To counter this, governments may need to consider more profound interventions like a Universal Basic Income (UBI) to ensure economic stability and provide a floor for citizens to stand on.

3. Incentivise Human-Centric Growth: The government can nudge the revolution positively. This includes funding research into human-AI collaboration, creating regulatory "sandboxes" for new AI-driven industries, and standardising the measurement of non-monetary goals like sustainability and social welfare. By making these targets legible and valuable, we can foster new economies that put our productivity surplus to work for the greater good.

The threat of rising inequality is particularly acute because AI-driven markets are subject to increasing returns, which naturally concentrate wealth. Proactive policy, potentially re-evaluating taxation of capital versus labour, is essential to ensure broad-based prosperity.

Conclusion: A Revolution of Choice

Generative AI is not just another technology; it is a turning point. Its ability to augment human intellect represents a new frontier. While the path ahead is fraught with uncertainty and the potential for severe disruption, it also presents a historic opportunity. We may have a genuine surplus of human cognitive ability for the first time.

The challenge is not to stop the wave, but to learn how to surf it. If we—as individuals, organisations, and societies—embrace adaptation, redesign work around human strengths, and consciously direct this new power toward our most pressing problems, the AI revolution could be the first to deliver not only unprecedented productivity but also a more sustainable and prosperous quality of life for all.

This is an updated version of my article published in the MBAEx Business Review.

You can find the original article here.