AI Gold Rush Can’t Stop Chip Sector’s Boom-Bust Cycle: Morningstar

The race for AI dominance has fueled a global chip frenzy — but Morningstar warns the industry’s next cyclical downturn may be on the horizon.

“Foundries and memory makers are exposed to the intense cyclicality of the semiconductor sector,” equity analysts at Morningstar cautioned in a report released on Tuesday.

The firm said a typical semiconductor cycle lasts about four years, with AI demand prolonging the current rally.

“We believe strong AI-related sales and investments are helping the sector to stretch the limits of an upcycle,” the analysts wrote.

Since ChatGPT’s breakout moment in late 2022, AI investments have surged, sending chipmaker valuations soaring.

While cloud giants like Microsoft, Amazon, and Meta are ramping up AI spending, the analysts said “the market is too upbeat on long-term AI investment growth.”

According to Morningstar’s analysis, semiconductor billings growth — a reliable gauge of industry health — has already started to slow, signaling the sector may be nearing its peak.

Outside of AI, sluggish smartphone and consumer electronics sales are dragging on demand for non-AI chips.

Related stories

Morningstar now sees AI spending peaking in 2025, with risks of a slowdown emerging in 2026 as macroeconomic risks intensify and consumer demand stays weak.

To be sure, cutting-edge AI chips remain scarce, but older memory products may see softer demand, they said.

While slowing growth would hit chipmakers, foundries are in a somewhat stronger position.

Morningstar pointed to Taiwan Semiconductor Manufacturing Company’s technological lead and massive US investments as buffers against risk. Still, even leaders like TSMC remain exposed to the cyclical swings that regularly sweep through the sector, they added.

Morningstar’s warning lands as investors wrestle with a broader question: how much AI will actually show up in corporate profits.

While a record share of S&P 500 firms mentioned AI on earnings calls in the second quarter, “the share of companies quantifying the impact of AI on earnings today remains limited,” Goldman Sachs said in a recent note.

The bank added that AI’s economic footprint may also be understated in government data, since semiconductor costs are often buried as intermediate inputs rather than fully captured in GDP data.