Intel Reportedly Eyes 14A2 Process Upgrade to Challenge TSMC and Samsung in the 1.4nm Era
Intel may be preparing a major update to its advanced semiconductor roadmap as competition in the 1.4nm-class chip manufacturing race intensifies. With TSMC and Samsung both pushing toward their next-generation foundry technologies, Intel is reportedly considering an enhanced version of its upcoming 14A process, currently referred to as 14A2.
The move would be aimed at strengthening Intel Foundry’s position against rival 1.4nm technologies expected later this decade. TSMC is preparing its A14 manufacturing platform, while Samsung is targeting mass production of its own 1.4nm-class node around 2029. Intel, meanwhile, is expected to introduce its 14A technology sooner, making it one of the company’s most important process nodes as it works to win more external foundry customers.
The rumored Intel 14A2 process is said to be a refined version of the standard Intel 14A node. One of the biggest changes could be the introduction of a dual-sided architecture for power delivery. Intel’s base 14A technology is already expected to use PowerDirect, a backside power delivery network designed to improve chip efficiency by moving power routing to the back side of the wafer.
With 14A2, Intel may go a step further by using power delivery from both the front and back sides of the chip. This dual-side approach could help improve performance, density, and power efficiency, all of which are becoming increasingly important as AI processors, data center chips, and high-performance computing products demand more advanced manufacturing techniques.
Another key improvement reportedly being considered for Intel 14A2 is a smaller M0 pitch. The standard Intel 14A node is expected to reduce M0 pitch to around 28nm, while 14A2 could push that figure down to approximately 21nm. A smaller pitch generally allows for greater transistor density, helping chip designers fit more logic into a smaller area.
Intel 14A is already expected to deliver a notable transistor density improvement of around 30% compared to previous technologies. If 14A2 moves forward with further pitch reductions and process refinements, it could offer another meaningful density boost. This would be especially valuable for advanced AI accelerators and next-generation CPUs, where every gain in performance per watt matters.
However, shrinking the pitch to 21nm is not without challenges. As wiring becomes denser, resistance can increase, potentially affecting power efficiency and signal integrity. There are also technical concerns around nano through-silicon vias, or nTSVs, which may not be ideal for handling the higher density requirements of such an aggressive process.
To address these issues, Intel is reportedly evaluating a composite structure. In this design, backside power delivery would remain the primary power source, while part of the power routing would also be assigned to the front-side metal layers. This hybrid approach could help balance density, efficiency, and manufacturability.
The use of double patterning is also expected to play a role in enabling these improvements. While double patterning adds complexity to manufacturing, it can help achieve tighter feature sizes and better density when used with advanced lithography tools. Intel’s 14A and possible 14A2 technologies are also expected to make use of High-NA EUV equipment, which is viewed as a critical step for future leading-edge chip production.
For Intel, improving utilization of High-NA EUV machines could also help boost manufacturing economics. These tools are extremely expensive, so increasing wafer output and yield per machine will be essential for making advanced nodes profitable.
The broader semiconductor industry is under intense pressure as demand for AI hardware, cloud computing infrastructure, and high-performance chips continues to rise. TSMC currently holds a dominant position in leading-edge foundry manufacturing, but its capacity is heavily booked by major customers. That has opened the door for other manufacturers, including Intel and Samsung, to compete for future chip orders.
Intel’s foundry strategy depends heavily on proving that its next-generation nodes can meet the needs of external customers. Technologies such as Intel 18A, 18A-P, 14A, and the rumored 14A2 will be closely watched by chip designers looking for alternatives in the advanced manufacturing market.
If Intel successfully executes its 14A roadmap and brings 14A2 to market with meaningful density and efficiency improvements, it could become a stronger competitor in the global foundry race. The company still has much to prove, but its push into backside power delivery, dual-sided architecture, and High-NA EUV manufacturing shows that it is aiming directly at the most advanced semiconductor technologies of the next decade.
The battle for 1.4nm-class chip production is shaping up to be one of the most important contests in the semiconductor industry. Intel’s rumored 14A2 process could be a key part of that fight, especially as AI and next-generation computing continue to drive demand for faster, denser, and more power-efficient silicon.
