MIT engineers are building 3D ‘high-rise’ AI chips for faster computing

MIT engineers are unveiling a groundbreaking method for creating multilayered 3D computer chips that could transform computing capabilities, particularly for artificial intelligence applications.

Technical breakthrough; MIT researchers have developed a novel approach to stack multiple layers of semiconducting materials directly on top of each other, creating what they call “high-rise” chips.

• The process operates at relatively low temperatures (below 400°C), ensuring the preservation of underlying circuitry while enabling vertical chip construction
• Using innovative “seed pockets” at mask opening edges, the team successfully grew single-crystalline 2D materials called transition-metal dichalcogenides (TMDs)
• The researchers demonstrated the technique by creating alternating layers of molybdenum disulfide and tungsten diselenide, two different types of TMDs

Performance implications; This new method represents a significant leap forward in semiconductor manufacturing capabilities and potential computing power.

• The technique effectively doubles the density of a chip’s semiconducting elements without requiring silicon wafer substrates between layers
• High-performance transistors and memory/logic elements can now be built on any crystalline surface, expanding beyond traditional silicon wafer limitations
• The advancement could enable laptop and wearable devices to achieve supercomputer-level performance with data center-scale storage capabilities

Research leadership and support; The project demonstrates strong institutional and industry backing, suggesting serious commercial potential.

• The research team is led by Jeehwan Kim, associate professor of mechanical engineering at MIT
• Both Samsung Advanced Institute of Technology and the U.S. Air Force Office of Scientific Research have provided support for the project
• Kim has established a company called FS2 to commercialize the technology

Future impact and market potential; The development of this 3D chip stacking technology could fundamentally alter the semiconductor industry’s trajectory and accelerate AI hardware development.

• The ability to stack chips without traditional limitations opens new possibilities for computing power optimization
• This breakthrough could lead to significant improvements in both logic and memory applications
• The technology’s commercial viability is reinforced by industry support and immediate steps toward commercialization

Looking ahead; While the technology shows immense promise, its successful implementation in commercial products will require careful navigation of manufacturing challenges and market dynamics, particularly as the semiconductor industry faces increasing pressure to maintain Moore’s Law-like advancement in computing power.