Samsung Plans to Use AI to Improve Its Chip Design and Production Processes to Better Compete with TSMC

Artificial intelligence (AI) has been making waves across almost every industry this year. The rise of ChatGPT has firmly cemented the technology in the public’s eye. Meanwhile, AI applications being used in the background for some time have become an attractive investment for businesses to show their investors and customers alike.  

Now, one of the world’s leading memory chip manufacturers is jumping on board. Samsung recently announced plans to revolutionize its production process with the help of AI and big data analytics. The cutting-edge approach will help Samsung compete with industry rivals, make its chipmaking process more efficient, and enhance wafer yields.

Harnessing the Power of AI

Samsung and TSMC are always in the conversation together where chip dominance is concerned. As such, the two are constantly in a back-and-forth battle to find an edge—no matter how small. Samsung believes its latest AI strategy will accomplish just that.  

The biggest improvement, the chipmaker hopes, will be its wafer yields during manufacturing. Samsung plans to use AI to analyze its processes and identify causes of unnecessary wafer loss as well as DRAM product defects. From there, the technology will give engineers ways to optimize the manufacturing process.  

Before the production line, Samsung also aims to use AI to automate the design of certain DRAM elements, chip materials, and chip packaging.  

The initiative is a partnership between the Samsung Advanced Institute of Technology (SAIT) and the company’s Device Solutions (DS) division. President of Samsung’s semiconductor segment, Kyung Kye-Hyun, is championing the project.  

Notably, the investment comes at a time when the DRAM market is in the midst of one of its worst slumps in history. While other chipmakers have scaled back their memory chip investments, Samsung is not only boosting its efforts but breaking ground in an innovative new space. With the right approach, this investment could pay off nicely when the market inevitably rebounds.  

Internal Development

Interestingly, Samsung won’t be relying on third-party algorithms to enhance its chip production. The company is currently working with Naver Corp. to develop its own generative AI platform as a competitor to ChatGPT in the corporate world.  

Internally, the platform will be used by Samsung’s DS division to develop an AI-based automated chip manufacturing system. The division will also create big-data algorithms to analyze its production lines and chip designs.  

Samsung plans to expand its innovative AI usage to other areas of its operations, as well, including the Device eXperience division, which is responsible for its smartphones and home appliances.  

In the meantime, the chip titan is recruiting AI experts to aid in the development of its innovative platform. Samsung is working with top universities as well as leading tech companies around the world to refine its strategy and tech.  

Competitive Requirement

Current chip manufacturing technology, including EUV lithography, is on the brink of its limits. Both Samsung and TSMC are working to achieve the 2nm node—and the winner will have a head start on cutting-edge chip orders. Samsung reportedly plans to commercialize its 2nm technology as soon as 2025. Meanwhile, TSMC started pre-production of its 2nm process in June and is rapidly approaching the first stage of trial production. Samsung hopes its AI-based approach will keep it competitive with TSMC while speeding up its implementation.  

Industry officials seem to agree the novel tactic is necessary as the scope of chip design rapidly expands beyond what human engineers can grapple with. One official said in a recent statement, “Chipmakers are increasingly adopting AI tech to identify the causes of defects that are nearly impossible for humans to identify or look for the most efficient ways to utilize the equipment.”  

Indeed, the chip industry is at a tipping point regarding AI—both in terms of production and demand. As the technology’s popularity soars, demand for powerful silicon to run massive AI data centers is climbing to a peak. Ironically, using AI to improve manufacturing and design processes is poised to be an integral part of meeting that demand over the next decade.  

As chipmaking becomes even more rigorous in the coming years, experts believe AI will be a major indicator of a company’s success. Samsung DS division chief Kyung says, “The gap between chipmaking companies that utilize AI and those that don’t will widen significantly.”  

For Samsung, the hope is that using AI will help it capture market share from its largest competitor, TSMC, all while improving its production process. Time will tell whether the bold strategy pays off and whether AI will truly reshape the chip industry.