Posted 8th May 2026

There’s a seductive idea at the heart of modern innovation: that technology can fix almost anything. Climate change? Build smarter grids. Food shortages? Optimise agriculture with AI. Urban congestion? Autonomous vehicles will sort it out. 

This mindset—often called tech solutionism—rests on a powerful belief that complex social and environmental problems can be solved primarily through better tools.

But what if that belief is part of the problem?

The Allure of Tech Solutionism

Tech solutionism thrives because it offers clarity in a messy world. It turns sprawling, systemic challenges into neat engineering problems. Instead of grappling with political will, cultural habits, or economic inequality, we get to focus on apps, algorithms, and devices.

Artificial intelligence sits at the centre of this movement. It promises efficiency at scale: optimising energy use, predicting climate patterns, and reducing waste in supply chains. These are real and meaningful contributions. AI can help us do things faster, smarter, and sometimes cleaner.

But the keyword is help, not solve.

The Hidden Costs of AI

AI is often framed as an invisible, almost magical force. In reality, it has a very physical footprint.

Training large AI models consumes enormous amounts of energy. Data centres require vast cooling systems, water usage, and continuous electricity—often sourced from non-renewable energy. The more we scale AI, the more we scale its environmental cost.

There’s also a resource story: rare earth minerals for hardware, global supply chains, and electronic waste. These impacts are rarely front and centre in conversations about “green AI,” but they matter.

So while AI may optimise emissions in one domain, it can quietly increase them in another.

When Technology Becomes a Distraction

Perhaps the bigger issue isn’t AI’s footprint—it’s what our reliance on it allows us to avoid.

Tech solutionism can act as a form of deferral. Instead of reducing consumption, we try to optimise it. Instead of rethinking growth, we attempt to make growth more efficient. Instead of behavioural change, we reach for automation.

For example:

• Smart thermostats may improve energy efficiency, but they don’t question why energy demand keeps rising.
• AI-driven agriculture can boost yields, but doesn’t necessarily address overproduction or food waste.
• Carbon capture technologies promise future fixes, potentially reducing the urgency for emissions cuts today.

In this way, technology can become a buffer between us and harder, more uncomfortable decisions.

What Real Sustainability Looks Like

Real sustainability is less glamorous. It’s slower, more political, and often inconvenient.

It involves:

• Reducing overall consumption, not just optimising it
• Designing systems that prioritise resilience over efficiency
• Addressing inequality and access, not just performance metrics
• Encouraging cultural shifts in how we live, travel, and consume

Technology—including AI—still has a role. But it works best as a supporting actor, not the main character.

For instance, AI can help monitor ecosystems, improve public transport systems, or model climate risks. But those tools need to sit within broader strategies that include policy changes, economic incentives, and collective behaviour shifts.

A Better Question to Ask

Instead of asking, “How can AI solve this?” we might ask:

• What problem are we actually trying to solve?
• Do we need to reduce, replace, or rethink the system entirely?
• Where does technology genuinely add value—and where does it distract?

These questions shift the focus from capability to necessity.

The Balance We Need

The goal isn’t to reject AI or innovation. It’s to rebalance our expectations.

AI is powerful, but it’s not neutral. It reflects the priorities of the systems that build and deploy it. If those systems prioritise growth above all else, AI will likely reinforce that pattern—even if it becomes more efficient along the way.

Real sustainability requires something deeper: a willingness to change not just how we do things, but why we do them.

And that’s something no algorithm can decide for us.