IBM has stolen a march on the semiconductor industry with a radical new chip architecture that resembles a block of flats rather than the traditional sprawling estates of logic gates. The 3D design, announced today at the International Electron Devices Meeting in San Francisco, stacks multiple layers of transistors vertically, promising to cram unprecedented computing power into a tiny footprint. For the UK, still smarting from the loss of its once-mighty chip manufacturing sector, this breakthrough offers a lifeline. The government’s recently unveiled National Semiconductor Strategy, which pledges £1 billion to bolster domestic fabrication capabilities, now has a concrete technological avenue to pursue.
The IBM ‘block-of-flats’ chip, officially called the Vertical Transport Field Effect Transistor (VTFET) design, breaks with the planar tradition that has dominated microchip manufacture since the 1960s. Instead of etching transistors side by side on a silicon wafer, IBM places them atop one another, with vertical ‘elevator shafts’ made of nanoscale wires connecting the layers. This allows for a much denser packing of components, reducing the distance data must travel and slashing energy consumption. In simulation, IBM claims a performance boost of up to 85% compared to the most advanced 5nm chips today.
For the British semiconductor ambition, timing is everything. The UK has lagged behind Taiwan, South Korea and the US in leading-edge chip production since the closure of many domestic fabs in the early 2000s. The government’s strategy, published last month, focuses on securing supply chains and investing in compound semiconductors and advanced packaging. But without a unique manufacturing technology, Britain would remain a niche player. IBM’s VTFET design could be the wedge that reopens the door. The company has already indicated it will license the technology to partners, and UK officials are quietly courting IBM to establish a pilot production line in the North East, where the semiconductor cluster around Durham and Newcastle still has skilled engineers from the old Fujitsu plant.
However, we must sound a note of caution. The history of chip manufacturing is littered with revolutionary designs that died in the lab. Vertical transistors have been tried before; Intel abandoned its own 3D stacking efforts in 2019 due to yield problems. IBM claims it has solved the heat dissipation issue by embedding tiny cooling channels between the layers, but mass production remains years away. Moreover, the geopolitical stakes are high. Duplicating such a complex process in the UK would require not just billions in investment but also a secure supply of ultrapure chemicals and advanced lithography tools from ASML, which is already constrained.
Yet the potential payoff is immense. A successful UK-based VTFET foundry could produce chips for everything from AI accelerators to IoT sensors, all while keeping design and manufacturing within friendly borders. For a nation that prides itself on innovation but has seen its industrial base hollow out, this represents a rare second chance. The user experience of society, after all, depends on resilient digital infrastructure. We cannot afford to be reliant on a single Asian island for the brains of our devices.
IBM’s announcement also carries a worrying Black Mirror subtext. With chips this powerful, the surveillance state becomes that much cheaper to run. The same vertical stacking that makes data centres more efficient also enables ever more invasive facial recognition and predictive policing. As the UK government celebrates this potential boon for its semiconductor strategy, it must also update its AI ethics guidelines. The hardware may be block-of-flats, but the governance must be high-rise ambition.
In the short term, investors will watch IBM’s stock tick up. In the medium term, the UK’s semiconductor task force will be working overtime on a feasibility study. And in the long term, this could be the moment Britain re-enters the big league of chip manufacturing. But only if the execution matches the vision. For now, we have a proof of concept. The real test will be whether we can turn this architectural drawing into a building we can live in.
