Why Is Your Phone's Price Rising? A Memory Crisis Rocks the Tech Industry

If you've noticed lately that the prices of smartphones, computers, and gaming consoles keep climbing, the reason isn't just general inflation, but a deep crisis striking one of the most important components in electronic devices: memory. After decades of falling memory costs that allowed devices to become more powerful without becoming significantly more expensive, the picture has completely reversed. Today, the rising price of RAM has become a primary driver of price increases across the entire tech sector.

A Crisis Beyond Just Artificial Intelligence

The common view blames artificial intelligence for this surge in prices. This technology not only raises energy bills and affects jobs, but also pushes up the prices of everything from phones to gaming consoles. But the reality is more complex; the crisis had been brewing before ChatGPT went viral, and then AI arrived with its voracious appetite for memory to amplify it in an unprecedented way.

Industry officials point to a fundamental shift in the nature of the challenge. For many years, memory manufacturers could simply increase production by cramming a greater number of cells onto each silicon wafer. But these technical gains began to shrink and take longer to achieve. Since 2021, Micron concluded that the problem had grown deeper: technical advancement alone was no longer enough to meet the rising demand, and manufacturers needed to process a larger quantity of wafers and build massive new facilities to do so.

A Market Collapse Followed by a Strong Rebound in Demand

What made matters more complicated was the earlier collapse of the memory market. Pandemic-era purchases of computers, tablets, and phones consumed future demand early, then consumer spending weakened, leaving manufacturers with surplus inventory and losses that pushed them to slow their expansion plans. When the market began to recover, generative AI unleashed a wave of demand that was larger and more memory-hungry than any manufacturer had anticipated.

A Highly Concentrated Market

The memory industry is characterized by exceptional concentration, with just three manufacturers controlling about 90% of the market, making the world dependent on a handful of companies to allocate limited production capacity between AI infrastructure and consumer devices. According to estimates, the shares were distributed as follows:

• Samsung: about 39% of the market
• SK Hynix: about 26%
• Micron: about 25%

Not All Memory Types Are Equal

The memory inside your phone or computer falls into two main categories. The first is Dynamic Random Access Memory (DRAM), which temporarily holds the information needed while opening apps, loading pages, and running programs. The second is stable storage memory (NAND flash), dedicated to storing photos and files over the long term.

Inside AI data centers, however, specialized processors rely heavily on a special type of DRAM known as High Bandwidth Memory (HBM). As experts explain, saying that data centers consume memory is not entirely accurate, because the type of memory used is different. HBM is produced by stacking chips on top of one another and using advanced interconnection and packaging techniques, allowing enormous amounts of data to be transferred with higher speed and efficiency.

Why Does AI Win the Priority?

HBM is harder to manufacture but more profitable to sell. AI chip makers and tech giants have an endless need for it and are willing to pay any price. More importantly, it consumes far greater manufacturing capacity—Micron estimates that producing a given quantity of HBM requires about three times the number of wafers needed to produce the same quantity of conventional DRAM.

The incentives lean more and more toward AI; rather than guessing how many phones or computers will sell a year from now, memory manufacturers can secure multi-year commitments from the world's richest companies. Samsung officials have indicated that they give priority to customers who guarantee steady future demand.

The pressure isn't limited to HBM—AI also raises demand for conventional DRAM, as phone and computer makers want to run smaller AI models directly on devices, which requires more advanced memory in larger quantities. All of this suggests that the memory price crisis may continue for years before finding its way to stability.

✦ بقلم فريق دروب أيديا