Alarm bells are ringing as Taiwanese NAND flash controller giant Phison Electronics once again claims a prolonged memory shortage. In a recent statement, Phison’s CEO Pan Chien-cheng, warned that AI demand could keep NAND flash and broader memory supplies tight for at least a decade, with severe shortages hitting in 2026.
Phison is a leading designer of controllers for SSDs and other flash memory solutions. The company sits at the heart of the industry shift and claims that in the coming two years, the biggest victims of hyper scaling of data centers will be consumer electronics.
In the next two years, the biggest victims will be consumer electronics, which is already unavoidable. For example, you may find it hard to believe that the main storage capacity for notebook computers in the second half of the year could regress to 256GB.
The crunch is already visible in pricing. NAND flash prices have doubled in just six months in some periods, with extreme examples like 8GB eMMC modules jumping from $1.50 to $20.
As long as AI applications continue, the demand for memory will not weaken.
Not everyone in the industry shares Pua’s decade-long timeline, but his views align with growing concerns. South Korea’s SK Group Chairman Chey Tae-won has similarly projected wafer shortages persisting until 2030.
However, there is another side to the AI story; one that could be pointing to a bubble.
Nearly half of AI data centers planned for 2026 are now facing delays or cancellation due to power shortages, permitting issues, and local oppositions. Many notable projects are being put on ice including a $1 billion Microsoft development in Ohio. Reason? Infrastructure bottlenecks. In total, $64 billion data center projects have been blocked or delayed, according to data center watch.
America’s AI data center boom might come to an end and that would mean no more memory shortages.
Consumers can expect higher prices for SSDs, laptops, smartphones, and other devices in the coming years, while the industry braces for potential consolidation. The only hope for consumers right now is a total collapse of the AI bubble because increasing manufacturing capacity will take years.
