The 10% Path and the 10X Path

Why using AI to cut costs may be the fastest route to irrelevance

For the last five decades, business leaders have learned what to do when a new technology arrives. These technologies usually enter through the IT department. They are evaluated, piloted, integrated into existing processes, and then used to make the organization a little faster, a little cheaper, and a little more efficient.

That playbook worked for mainframes, PCs, databases, the internet, cloud computing, and mobile technology. So it is entirely understandable that many leaders are now applying the same approach to AI. But AI is not just another technology in the stack. It changes what an organization can produce, how quickly it can learn, and how far it can scale.

Leaders face two very different paths. And only one leads to good long-term outcomes. Let me explain.

The familiar path: optimize the past

The first path is the one most organizations already know. Leaders look at their current operations, identify expensive activities, and ask where AI can replace labor, reduce costs, or automate existing tasks. The goal is usually framed in terms of efficiency: How can we do the same work with fewer people, less time, and a smaller budget?

AI as just another efficiency tool

This path is attractive because it fits the way companies already operate. Budgets, headcount, capital equipment, and organizational capacity are treated as fixed constraints, and leaders are rewarded for squeezing better results from the resources they have. That’s certainly the world I grew up in at Intel.

With this mindset, AI becomes another tool for tightening the machine. A customer service team might use AI to handle more inquiries with fewer (human) agents. A finance department might automate reporting. A manufacturer might reduce the number of people inspecting products or managing inventory.

There is nothing inherently wrong with any of this. Every organization should remove waste, automate repetitive work, and improve productivity. The problem is that efficiency alone is not a strategy.

The limits of incremental improvement

At best, this approach may produce a 10-20% improvement in your results. Costs will fall, margins might improve temporarily, and quarterly results will look healthier. Backs will be slapped, management bonuses will fly, and some people will get laid off in the process.

But the organization remains fundamentally the same. It sells the same products, serves the same customers, follows the same processes, and competes in the same way. You've just used AI to optimize the past.

A race to the bottom

Now imagine two competing companies, both following this path. They have similar access to capital, models, software, cloud infrastructure, and automation tools. Both identify the same kinds of jobs to eliminate and the same expenses to reduce. One company cuts 8% of its workforce and the other cuts 10%.

Before long, both companies have stripped away many of the things that once made them distinctive. They compete more aggressively on price, because price is one of the few levers they have left. This is the corporate equivalent of removing every comfortable seat, free snack, and friendly employee from an airline and then wondering why customers feel no loyalty to the brand. Both companies may become more efficient. Neither becomes more valuable. Employees lose jobs, customers receive a thinner experience, and shareholders own companies that are leaner but increasingly interchangeable.

It is a race to the bottom—and there is rarely much of a prize for arriving first.

The second path: invent the future

Now imagine a third company taking a different route. Its leaders still care about efficiency, but they ask, “What could we accomplish if intelligence and labor were no longer our primary constraints?”

In my experience, asking this question is the foundation of a worthwhile AI transformation strategy.

From labor substitution to labor amplification

Instead of using AI primarily for labor substitution, this third company uses it for labor amplification. It gives people access to teams of AI agents that can research, analyze, design, simulate, communicate, monitor, and execute work at enormous scale.

A product manager might test hundreds of product concepts using simulated markets. A sales team might serve thousands of prospects with highly personalized outreach rather than concentrating only on the largest accounts. An engineer might explore millions of possible design configurations instead of three. A healthcare organization might evaluate treatments, predict patient needs, and personalize care with greater precision.

The company doesn’t ask people to work faster, it extends their abilities and gives them new capabilities that were previously impossible. The result: they can now all deliver far more and create much higher impact for the company.

Scaling knowledge work and physical work

As physical AI matures, AI agents may also coordinate with physical systems and increasingly capable robots. Knowledge work and physical work both begin to scale, while people remain responsible for setting goals, exercising judgment, building trust, protecting quality, and deciding what work is worth doing. More on this in my post on the role of people in an AI-first company.

This approach changes the economics of the enterprise. Traditional companies scale by adding people, facilities, equipment, and management layers. Growth usually increases cost in roughly the same direction as revenue. But AI-first companies can scale intelligence and execution much more rapidly, without a commensurate impact to costs. Agents and robots cost money, but not as much as the people you’d need to hire to scale operations. And with time, improved algorithms, streamlined open weight models, and volume economics will yield cheap robots and agents that operate at near the cost of electricity.

When marginal costs collapse

For this third company, as output expands, marginal costs collapse. The hundredth customer, design, analysis, or product variation may cost dramatically less to produce than the first. The company is no longer focused on reducing inputs but on expanding outputs: serving more customers, creating more new products, entering new markets, and delivering more value to more people.

This is the difference between improving a horse-drawn carriage and inventing the automobile. One path creates a slightly better version of what already exists; the other changes what’s possible.

What happens when the three companies compete?

Now put all three companies in the same market. The first two have automated aggressively, reduced headcount, and cut costs. They may be 15% more efficient than they were before. The third company has amplified its people, expanded its reach, launched new services, created faster learning loops, and built an organization capable of operating at a very different scale. It may be able to serve ten times as many customers, test one hundred times as many ideas, or create twenty times as much output with a more moderate increase to its traditional cost base. This leverage should be at the heart of every AI transformation strategy.

One company is playing a different game

Which company is more likely to win? The answer isn’t difficult. While the cost-cutters fight over slices of the existing pie, the third company is building a much larger bakery and expanding into entirely new markets. Ironically, the company pursuing growth may also become the lowest-cost competitor. But its lower costs come from scale, automation, learning, and dramatically greater output—not simply from removing people. Cutting labor can improve a spreadsheet once. Building a self-improving system can compound advantage year after year.

From managing scarcity to creating abundance

The real challenge for leaders isn’t technological, it’s psychological. Most executives have spent their careers learning how to manage scarcity. There’s never enough budget, time, talent, capacity, or information to deliver on their ambitions. Great leaders became great by allocating those scarce resources wisely.

But AI introduces a different management challenge: what happens when intelligence becomes abundant? When every employee can access extraordinary analytical, creative, and operational capacity. What happens when digital labor can be replicated quickly, knowledge can be applied continuously, and an organization can learn from every interaction? This represents an entirely new mindset.

A new leadership mindset

Leaders must begin thinking less like custodians of constraints and more like architects of abundance.

And that’s not an easy transition. When you’ve won the Super Bowl thirty times using the same playbook, it’s difficult to accept that the game has changed. Habits that made you wildly successful feel like wisdom, even when they might now be limitations. My advice: you don’t need to abandon discipline, efficiency, or financial responsibility. But stop treating them as the destination.

The question leaders should ask

Use AI to remove waste. Automate repetitive work. Lower costs where it makes sense. But don’t stop there. Ask what your organization can do now that it could never do before. Ask how you could use AI to become the company you’ve long dreamed of building, if only you weren’t constrained. And choose the mindset and the path that’ll take you there.

One path uses AI to preserve your existing business for a little longer. The other uses AI to build the business that replaces it. Leaders must now choose carefully, because only one of those paths leads to the future. The other probably leads to long-term irrelevance.

Choose carefully: AI offers new levels of potential that will dramatically favor only the bold.

Let me know how I can help,
Steve

Steve Brown is an AI futurist, global keynote speaker, and author. He advises Fortune 100 companies and global brands on AI strategy and transformation, helping them build AI-first organizations and stay ahead of the forces reshaping business, the economy, and society. He has delivered hundreds of high-impact keynotes across five continents, translating complex technologies into clear, practical action for leaders. Book him to speak at your next event here.

Steve’s latest book is “The AI Ultimatum: Preparing For a World of Intelligent Machines and Radical Transformation.” Get his book here, and learn more about Steve at www.stevebrown.ai.