Samsung Combines AI and Quantum Computing to Produce the Best Chips Ever, at a Lower Cost

Samsung is working on a new platform for semiconductor production that combines quantum computing and artificial intelligence to simulate the lithography phase, considered the most important step in the entire manufacturing process. According to reports from the Seoul Economic Daily, the goal is to reduce processing times and costs while simultaneously increasing the density and yield of the chips.

Lithography represents the first phase of semiconductor manufacturing. In this step, high-precision scanners transfer the design of integrated circuits onto the wafer. For the most advanced production nodes, systems based on extreme ultraviolet light (UV lithography) come into play, which are essential for creating increasingly smaller transistors.

Samsung is developing algorithms dedicated to simulating lithography that will run on quantum computing systems. The aim is to predict the behavior of the manufacturing process before the physical creation of the wafers, thereby optimizing two fundamental parameters of the semiconductor industry. The first is density, meaning the number of transistors that can be integrated into a given surface area of the chip. The second is yield, referring to the number of perfectly functioning chips obtained from each wafer. Improving both of these aspects helps increase production efficiency and contain manufacturing costs.

Once the simulation is complete, artificial intelligence comes into play, tasked with identifying any errors and automatically correcting anomalies that emerge during the lithography simulation. In this way, Samsung aims to refine the subsequent processing stages, including etching.

The project is being developed by Samsung SDS, the digital division of the group. According to available information, the company has already completed the necessary algorithms and plans to verify their functionality through a proof of concept in the coming year. If successful, the software could be transferred to Samsung Electronics for use in semiconductor production activities.

The use of artificial intelligence in chip production is not entirely new for the industry. Several companies are already using AI technologies to optimize production process development. In Samsung's case, however, it lies in the integration of quantum processing in lithography simulation. Beyond increasing density and yield, Samsung's solution should also ensure a higher quality of chips, improving the so-called "silicon lottery."