Samsung Electronics and Broadcom have entered into an agreement Memorandum of understanding (Memorandum of Understanding) in order to enhance their cooperation in the fields of memory, foundry manufacturing and advanced packaging. They expect to spend more than $200 billion as a result of their partnership through 2030. The agreement comes at a time when many AI chip makers are trying to secure their supplies of high-bandwidth memory (HBM), one of the major challenges in the industry.
The agreement was announced at the Artificial Intelligence Summit held in San Francisco where South Korea participated President Lee Jae-myung Representatives of prominent artificial intelligence companies participated. The South Korean presidential office considered Broadcom’s cooperation as one of the initiatives to boost the local artificial intelligence semiconductor industry.
While Samsung acknowledged the expected financial outlook and size of the partnership, the company and the South Korean president’s office declined to say how many billions they would save each year, what the prices would be, when shipments would be scheduled, or how the amount would be split across memory, foundry, and advanced packaging. Since there is no order or purchase agreement, the estimated value of $200 billion would merely represent a long-term strategy, not actually identified revenues.
Why is buying memory from Broadcom important outside of the two companies?
Broadcom doesn’t make memory chips, but it has become one of the world’s largest designers of custom AI accelerators. Samsung said the broader partnership will support Broadcom’s next generation of AI chips through advances in HBM, foundry manufacturing and advanced packaging. Broadcom is really evolving Google Tensor Processing Units (TPUs) and MTIA AcceleratorsBoth require large amounts of HBM.
This is why the agreement has the potential to create waves in the AI sector. HBM remains one of the most constrained components in terms of availability in AI hardware. The Samsung-Broadcom deal could secure full manufacturing capacity that other chip developers may want access to as well.
It is already clear that demand is rising. According to Seoul Economic Daily, SK Group President Chi Tae-won He said Broadcom has been constantly ordering what he calls a staggering amount of memory chips, highlighting how hyperscale AI customers are working hard to make sure they are able to buy enough supplies for the future.
Data from industry supports this claim. TrendForce Q1 2026 report on HBM technology showed SK Hynix Samsung remains at the top of the market, while Samsung recorded the fastest recovery due to increased shipments of HBM3E followed by the marketing launch of HBM4. An update from TrendForce, issued later, said that Samsung had passed the validation HBM4 It began shipping them before any other supplier, which strengthens its position in the face of the growing demand for artificial intelligence.
The deal arrives as memory supplies remain limited
Timing may be just as important as agreement.
The cost of memory has been growing throughout the year, as demand for AI servers exceeds the supply of these servers, and experts do not expect this situation to change anytime soon.
On July 3, forecasting firm TrendForce predicted that contracts Dirham chips In the third quarter of 2026, they will rise by 13-18% compared to the previous quarter, while NAND chip prices are expected to jump by 10-15%, confirming that the DRAM chip market is very tight.
According to SemiAnalys, the industry has entered into what is called “Silicon deficiency stage“, during this period the production of advanced logic and memory is unable to keep up with funding for AI infrastructure. In this scenario, companies willing to make long-term commitments secure more certainty regarding future supplies while manufacturers find themselves locked into long-term contracts with customers due to their very limited capabilities.”
Samsung’s attempt to regain its place in artificial intelligence memory
Samsung has struck an agreement with Broadcom amid efforts to strengthen its market position in AI memory after losing leadership in HBM to SK hynix. The company said that the partnership with Broadcom brings together its expertise in memory, foundry production and advanced packaging to create a comprehensive platform for AI customers.
Samsung had revealed a few months ago that it had begun mass production of the HBM4, and then on May 29, it released what it claimed was the first 12 layers HBM4E Samples available for big customers. Samsung has confirmed that the latest addition to its memory lineup can achieve transfer speeds of up to 16 Gbps, in addition to 3.6 Tbps of bandwidth per packet, targeting future AI accelerators.
Its foundry business also gained momentum. Samsung Foundry has secured Tesla’s AI5 and AI6 software alongside TSMC and entered Nvidia’s data center supply chain, SemiAnalogy reported.
If Broadcom’s note eventually turns into firm production orders, Samsung will add another high-profile AI customer to its expanding semiconductor portfolio. The South Korean presidential office also said that Korean and global technology companies announced $950 billion in semiconductor cooperation during the AI Summit, making the Broadcom agreement one of the event’s largest strategic commitments.
Don’t just read cryptocurrency news. Understand that. Subscribe to our newsletter. It’s free.
