Can We Boost Productivity Without Relying Only on AI?

Each time fears around artificial intelligence (AI) stealing jobs surge, there’s an equal wave of reassurance from tech optimists. They argue that AI isn’t coming for our livelihoods, it’s here to help us work smarter. 

Tools like autonomous AI agents, as envisioned by Microsoft’s Satya Nadella, promise a future where we simply state our goals and the software handles the rest. 

On the surface, it’s a dream scenario for productivity. But is efficiency alone the answer to our productivity problems?

When technology isn’t enough

We’ve seen this movie before. Over the last few decades, we’ve added faster computers, smarter software, and global connectivity to our workplaces. Yet productivity growth particularly in advanced economies has slowed rather than surged. 

In the 1990s, labour productivity grew around 2% a year. Today, it’s closer to 0.8%. Even in once-booming China, output per worker has flatlined.

So, what’s the missing ingredient?

More tools, fewer breakthroughs

One major issue is that we’ve mistaken faster execution for deeper innovation. The combination of computers and the internet was supposed to spark a golden era of human progress by democratizing access to information and connecting talent globally. 

Instead, we’ve seen a steady drop in research output. Scientists today produce fewer groundbreaking ideas per dollar than they did in the 1960s.

Why? Because creativity doesn’t thrive on speed alone. It thrives on focus, patience, and the freedom to explore. Newton spoke of holding a single problem before him until insight “dawned slowly.” 

Steve Jobs was famous for saying innovation means “saying no to a thousand things.” In other words, more projects don’t mean better ideas they often mean diluted thinking.

The risk of AI as a shortcut

AI, particularly large language models (LLMs), are exceptional at automating routine tasks. But they often mirror the consensus of the data they were trained on. If fed outdated or mainstream knowledge, they simply repeat it with confidence not question it. For instance, an AI trained before Galileo would reinforce a geocentric universe; it wouldn’t challenge it.

Even AlphaFold, arguably AI’s most significant scientific success so far, required major human insight to be built in the first place. As its creators at Google DeepMind admit, we’re still a long way from machines that can rival human cognitive leaps across all domains.

Efficiency vs. Originality

A recent survey of over 7,000 knowledge workers showed that heavy users of generative AI saved about 3.6 hours a week on email tasks. 

That’s impressive until everyone’s using AI to write emails, and the volume of communication increases again. Efficiency gains often cancel out once the playing field evens out.

What’s more lasting are breakthrough innovations discoveries that change industries and lives. These don’t emerge from doing the same things faster, but from doing new things entirely. 

Antibiotics, jet engines, rockets they weren’t the result of improved workflows. They were bold leaps into the unknown.

The human edge

To truly revive productivity, we must shift our focus from just speeding up tasks to creating space for original thinking. That means embracing riskier ideas, allowing people the autonomy to explore, and building institutions that reward invention not just output.

AI can be a powerful tool in that journey. But it’s not the destination. If we use it simply to clear to-do lists faster, we’ll soon hit a wall. If we use it to rethink what we’re capable of doing, then we stand a real chance at solving the productivity puzzle.

In the end, boosting productivity isn’t about choosing between humans or AI it’s about making sure the human drive to discover stays at the heart of whatever tools we build next.