The broad semiconductor shortage that disrupted industries from 2020 through 2023 is over for most chip categories. However, a new and different supply crunch is emerging: memory chips used in everything from phones to PCs are being squeezed by massive demand from AI data centers, and that pressure is expected to last through 2026 and into 2027.
How the Original Shortage Ended
The global chip shortage that began during the pandemic was driven by a collision of factory shutdowns, surging demand for electronics, and a supply chain that couldn’t flex fast enough. The U.S. Government Accountability Office places the shortage window at 2020 through 2023, and the federal government has since funded projects aimed at strengthening domestic production at every stage, from raw materials to final packaging.
By mid-2024, most categories of chips had returned to normal availability. Automotive semiconductors, one of the hardest-hit sectors during the crisis, saw average lead times drop to about 14 weeks in the second quarter of 2024. That’s roughly in line with pre-pandemic norms of under 17 weeks. The one exception: automotive processors still averaged above 20 weeks, a lingering hangover from the earlier crunch but far better than the 40-plus-week waits that plagued automakers in 2021 and 2022.
Memory Chips Are the New Bottleneck
While the old shortage has faded, a new one is taking shape. The memory and storage chips inside your phone, laptop, and tablet are increasingly competing for factory capacity with the chips powering AI infrastructure. Data centers building out systems for artificial intelligence need enormous quantities of the same types of memory (DRAM and NAND flash) that go into consumer devices, and that competition is straining supply.
IDC expects these memory supply challenges to persist throughout 2026 and likely well into 2027. The structural cause isn’t going away: AI infrastructure demand keeps growing, and it draws from the same pool of manufacturing capacity that serves consumer electronics. Unlike the pandemic-era shortage, which was triggered by temporary disruptions, this one is rooted in a fundamental shift in how memory chips are being consumed globally.
What This Means for Phones and PCs
The practical effects are already showing up. As memory prices climb, device makers face a difficult choice: absorb higher component costs, raise prices, or ship products with less memory than consumers have come to expect. Many are doing all three.
IDC forecasts the worldwide PC market will decline 11.3% in unit shipments in 2026, even as revenue grows 1.6% because each unit costs more. The smartphone picture is worse: a projected 12.9% decline in units shipped, with revenue dipping slightly by 0.5%. In both markets, the sticker price goes up while fewer devices actually sell. Some manufacturers, particularly smaller ones, are struggling to secure enough memory supply at prices they can afford.
This dynamic is expected to benefit the largest global device makers, who have more purchasing power and stronger supplier relationships. IDC anticipates meaningful market share shifts toward these bigger players over the course of 2026. If you’re shopping for a phone or laptop, you may notice fewer budget options and some models offering less storage or RAM than their predecessors.
Chip Prices Overall Have Stabilized
Outside the memory segment, semiconductor pricing has largely normalized. According to the U.S. Bureau of Labor Statistics, producer prices for semiconductor manufacturing rose 2.2% in 2024, a modest increase after a slight dip of 0.1% in 2023. Import prices were essentially flat in 2024, following a 3.8% decline in 2023. Export prices edged down 0.2%.
These are not the wild swings of 2021 and 2022, when spot prices for some chip categories doubled or tripled. The overall pricing environment is calm, which reinforces that the general shortage is behind us. The pain is concentrated in memory, not in the logic chips, microcontrollers, and sensors that go into cars, appliances, and industrial equipment.
Why Supply Isn’t Bouncing Back Quickly
Building new semiconductor manufacturing capacity takes years. A cutting-edge chip fabrication plant costs tens of billions of dollars and typically takes three to five years from groundbreaking to volume production. Memory fabs face similar timelines. Even with significant government investment flowing into new U.S. facilities, the added capacity won’t come online fast enough to relieve the current memory squeeze in 2026.
The challenge is compounded by the nature of AI demand. Training and running large AI models requires specialized chips packed with high-bandwidth memory, and every major cloud provider and tech company is racing to build out capacity simultaneously. That creates a demand spike that even well-supplied markets would struggle to absorb gracefully.
For most consumers, the era of empty car lots and year-long waits for game consoles is over. But if you’re buying a new phone or computer in 2026, expect higher prices, potentially less memory in mid-range devices, and fewer choices from smaller brands. The semiconductor world isn’t in crisis the way it was in 2021, but it isn’t fully at ease either.