Samsung and Broadcom signed a $200 billion memorandum of understanding in late July 2026, one of three mega-deals reshaping chip and memory supply chains that month.1 Nvidia and SK Hynix followed with a $500 billion memory agreement.2 South Korea and the US then unveiled a roughly $950 billion AI summit package.3 All three landed within days of each other.
The clustering signals hyperscalers and sovereign capital moving into the same supply chains simultaneously. Memory, advanced packaging, and AI accelerator production are the common targets.
Consolidation is also reshaping the sector's edges. Skyworks and Qorvo announced a merger with combined leadership meant to align the two RF/analog chipmakers. "Today's announcement reflects the strong partnership that has shaped our integration planning efforts from the very beginning," said Bob Bruggeworth.4 He said he is confident the named leaders will foster collaboration across the merged teams.
EDA leaders Cadence, Synopsys, and Siemens are embedding Nvidia's agentic AI tools into chip design workflows, extending the AI buildout upstream into design software itself.5
Traditional semiconductor names are posting record results alongside the capital wave. Amkor and Allegro both reported record figures. Vicor Corporation posted trailing twelve-month revenue of $471.7 million and net income of $136.7 million, with a market capitalization of $11.3 billion as of the July 6, 2026 close.
Public markets are pricing in the buildout. The iShares Semiconductor ETF (SOXX) is up roughly 79% year to date and about 152% over one year.1
Together, the Samsung-Broadcom MOU, the Nvidia-SK Hynix deal, and the Korea-US package point to a supply chain being rebuilt around hyperscale and state-backed capital rather than conventional corporate spending cycles. The scale and timing — three deals in the same month — mark a shift in how AI infrastructure gets financed.
