Future AI chips could be built on glass
View original at technologyreview.comMIT Technology Review - Ai Research Title: Future AI chips could be built on glass Date: 2026-03-13 09:00 Source: https://www.technologyreview.com/2026/03/13/1134230/future-ai-chips-could-be-built-on-glass/ <p>Human-made glass is thousands of years old…
O que extraímos desta fonte
The claims Via News extracted from this document. We point to the source; we don't replace it.
Glass thermal stability could allow engineers to create 10 times more connections per millimeter than organic substrates
60% confidenceGlass can be made 5,000 times smoother than organic substrates, eliminating defects that can arise as metal gets layered onto semiconductors
60% confidenceIntel realized about a decade ago that there would be limitations with organic substrates
60% confidenceThe benefits of glass core substrates are undeniable and will drive the industry to make this happen sooner rather than later
60% confidenceGlass holds enormous potential for the future of energy-efficient AI compute because light-based systems could move signals with far less energy than power-hungry copper pathways
60% confidenceIn early testing days, hundreds of glass panels got cracked every couple of days at Intel
60% confidenceGlass substrates unlock the ability to keep scaling package footprints without hitting a mechanical wall
60% confidenceMultiple large manufacturers including Samsung Electronics, Samsung Electro-Mechanics, and LG Innotek have significantly accelerated their research and pilot production efforts in glass packaging over the past year
60% confidenceGlass can be made 5,000 times smoother than organic substrates, which would eliminate defects that can arise as metal gets layered onto semiconductors
60% confidenceIntel's R&D teams are reliably fabricating glass panels and producing test chip packages, a significant improvement from early testing days when hundreds of glass panels got cracked every couple of days
60% confidenceMultiple large manufacturers including Samsung Electronics, Samsung Electro-Mechanics, and LG Innotek have significantly accelerated their research and pilot production efforts in glass packaging over the past year, suggesting the glass substrate ecosystem is evolving from a single early mover to a broader industrial race
60% confidenceThis is not the first attempt to adopt glass in semiconductor packaging, but this time the ecosystem is more solid and wider and the need for glass-based technology is sharper
60% confidenceThe benefits of glass core substrates are undeniable and will drive the industry to make this happen sooner rather than later
60% confidenceAbsolics has led the way in commercializing glass substrates
60% confidenceGlass substrates unlock the ability to keep scaling package footprints without hitting a mechanical wall
60% confidenceGlass thermal stability could allow engineers to create 10 times more connections per millimeter than organic substrates
60% confidenceAbsolics has led the way in commercializing glass substrates
60% confidenceThe semiconductor market for glass could grow from $1 billion in 2025 to as much as $4.4 billion by 2036
60% confidenceIntel realized about a decade ago that there would be limitations with organic substrates
60% confidenceIntel's designers can stuff 50% more silicon chips into the same package area with glass substrates, improving computational capability
60% confidenceAs AI workloads surge and package sizes expand, the industry is confronting very real mechanical constraints that impact the trajectory of high-performance computing, with warpage being one of the most fundamental issues
60% confidenceThe semiconductor market for glass could grow from $1 billion in 2025 to as much as $4.4 billion by 2036
60% confidenceThis is not the first attempt to adopt glass in semiconductor packaging, but this time the ecosystem is more solid and wider, and the need for glass-based technology is sharper
60% confidenceAs AI workloads surge and package sizes expand, the industry is confronting very real mechanical constraints that impact the trajectory of high-performance computing, with warpage being one of the most fundamental issues
60% confidenceGlass holds enormous potential for the future of energy-efficient AI compute because a light-based system could move signals around with far less energy than power-hungry copper pathways
60% confidence