Andrew Bolwell's avatar
Posted byAndrew Bolwell

The global technological landscape is undergoing a transformation driven by several converging forces.

These emerging innovations represent not just incremental improvements, but potential paradigm shifts that could fundamentally alter industries, economies, and societies.

While attention often focuses on established trends, it’s the subtle indicators — the weak signals — that offer the most valuable insights into future disruptions. These early indicators, though not yet mainstream, carry significant implications for strategic planning and competitive advantage.

What trends are we seeing?

The tech landscape is constantly shifting, composed of established trends as well as weak signals. Some trends may be game changers while others are foundational breakthroughs, along with a few wild cards still in their nascent phase. Here are some of those trends we are watching closely.

  • Artificial General Intelligence (AGI)
  • Humanoid robots
  • Quantum computing
  • 6G Networks and hyperconnectivity
  • Advanced compute
  • Nuclear energy
  • Biotechnology and synthetic biology
  • Next generation energy storage
  • Clean tech
  • Space tech

Today, I’m diving into two of these trends: advanced compute and clean tech and energy.

Advanced compute

Driven by AI, the increasing demand for computing power is colliding head-on with energy constraints that challenge this infrastructure build out. The result? We will need advancements in both energy efficiency and energy creation to keep up. Innovations such as microfluidic-cooled chips and composable computing architectures are driving computing efficiency and performance to new heights.

At the same time, data centers, cloud computing, and other high-performance applications are placing immense pressure on the global energy supply, pushing the need for cleaner, more reliable energy sources. We’ve seen numerous clean tech products and services focused on energy efficiency and sustainable reuse, and you can expect this trend to accelerate.

The question is: How quickly can we implement solutions that strike a balance between technological growth and sustainability? The interaction between computing and energy production is becoming one of the most critical challenges and opportunities of the modern era.

Compute is getting smarter — and hungrier

Computing advancements are reshaping industries, bringing both efficiency gains and new challenges.

The evolution from massive supercomputers to high-performance, compact chips makes processing power more accessible and scalable, enabling more sophisticated AI models, data analysis, and automation.

These innovations drive digital transformation across sectors, from healthcare to finance, but they also come with a steep energy demand. AI-driven applications, especially large-scale models, require immense computing power, leading to a surge in electricity consumption and an urgent need for clean energy solutions to support this growth.

Today, data centers account for 1% to 2% of overall global energy demand, similar to what experts estimate for the airline industry. When costs related to delivering AI to the world is factored in, that figure is expected to hit 21% by 2030.

Moreover, the International Energy Agency projects that data centers will use 945 TWh of electricity in 2030, roughly equivalent to the current annual electricity consumption of Japan according to Nature.

New energy sources, storage, and compute power

This is resulting in looking at new energy sources. Including a revival of nuclear power, which today accounts for nearly 10% of the world’s electricity but could grow significantly in the coming years due in part to its low-carbon footprint.

That’s why Small Modular Reactors (SMRs) from startups such as Nuscale and TerraPower are stepping into the spotlight as a potential answer to powering AI-driven data centers, offering a steady and reliable energy source that can be deployed more flexibly than traditional nuclear plants. These reactors can generate consistent, carbon-free electricity, making them an attractive option for reducing the environmental footprint of high-performance computing. SMRs can generate up to 300 MW per unit, which is significant when considering that a typical hyperscale data center requires 20–50 MW of power capacity, making them an ideal power source for AI-driven supercluster data centers.

Similarly, looking to other sustainable energy innovations in solar, wind, and gas will be needed for AI to advance. But it’s not just how we create power but also store and distribute it that will be key to new compute models. Solid-state batteries, or SSBs, could play a crucial role in grid storage applications needed to power the future computing needs of AI. While initially being developed for electric vehicles, these next-generation energy storage solutions have the potential to support gri-dscale applications, helping bridge the gap between fluctuating renewable energy sources and computing technology’s increasing power demands. In fact,the solid state battery market is expected to grow at a compound annual growth rate (CAGR) of 33%, with commercialization efforts ramping up.

Advanced Compute Breakthroughs

The energy challenge is also driving innovation in computing itself. New processor architectures designed specifically for AI workloads are dramatically improving performance while reducing energy consumption.

Neuromorphic computing, which mimics the efficiency of the human brain, shows promise for reducing energy requirements by orders of magnitude compared to traditional computing approaches. Research from Intel’s Neuromorphic Research Lab demonstrates that neuromorphic systems can be up to 1,000 times more energy-efficient than conventional architectures for certain AI workloads.

Quantum computing developments could revolutionize how we approach certain computational problems, potentially solving in seconds what would take today’s supercomputers years, all with a fraction of the energy.

These compute breakthroughs, paired with energy innovations, will be essential to sustainable AI growth.

The Power Players

The compute and clean energy race is full of power players. Here’s where innovation is heating up:

Advanced Compute

Clean Tech

Nuclear Energy

Next-Gen Energy Storage:

Unexpected outcomes

The AI energy equation could take unexpected turns in coming years:

What if AI algorithms emerge that drastically reduce computational requirements? Some researchers are exploring “small language models” that deliver impressive results with far less computing power, similar to how the human brain achieves remarkable efficiency.

IBM, Google, Microsoft, and OpenAI have all recently released small language models (SLMs) that use a few billion parameters — a fraction of their bigger LLM counterparts.

Geopolitical implications could be profound if certain nations successfully implement nuclear or other sustainable energy solutions for AI infrastructure while others remain reliant on fossil fuels. Countries that solve the energy puzzle could gain significant advantages in the AI arms race, potentially reshaping global power dynamics.

Could success in one domain–either compute efficiency or clean energy–accelerate the other in a virtuous cycle? Or might we face a scenario where breakthroughs in AI capabilities consistently outpace our ability to power them sustainably?

A global shortage of critical minerals like lithium could emerge as a limiting factor, constraining the growth of both advanced computing and clean energy technologies.

These minerals are essential for producing high-performing batteries, semiconductors, and energy storage systems–components that power AI data centers, renewable energy grids, and more.

AI’s promise could become a paradox if we don’t solve the energy storage issue. We will have smarter tools powered by unsustainable systems.

Moving forward

As computing power and energy needs evolve, the intersection of advanced compute and clean energy will shape the next wave of technological and environmental progress.

The race to develop sustainable, high-performance computing solutions is accelerating, and innovations in energy efficiency, storage, and nuclear technology will define the next decade.

Will we rise to meet the energy demands of intelligent machines? Or will innovation outpace our ability to power it? I want to hear your thoughts on what breakthrough or roadblock you see defining the next decade.