Google's new AI memory-compression algorithm, TurboQuant, has sent a chill through the semiconductor sector, raising questions about the sustainability of the memory supercycle.
Back
Google's new AI memory-compression algorithm, TurboQuant, has sent a chill through the semiconductor sector, raising questions about the sustainability of the memory supercycle.

Micron Technology (NASDAQ:MU) shares fell 5% to $339 after Google unveiled TurboQuant, an AI memory-compression algorithm that threatens to reduce demand for the high-bandwidth memory chips that have fueled a 289% rally in the stock over the past year.
"This closes the inference cost gap," Morgan Stanley analyst Joseph Moore said, arguing that more efficient memory use will ultimately spur more intense and complex AI workloads rather than curb overall hardware demand.
The selloff, which also hit equipment maker Lam Research (NASDAQ:LRCX) with an 8.7% drop, was triggered by Google's claim that TurboQuant can reduce the memory size for large language models by at least six times without sacrificing accuracy. This comes as Micron's high-bandwidth memory (HBM) capacity is reportedly sold out for all of 2026, with the company projecting a 40% compound annual growth rate for the HBM market through 2028.
The conflict between TurboQuant's efficiency promise and the insatiable demand for AI hardware is the new battleground for semiconductor investors. While Micron trades near all-time highs, the market is now forced to price in the risk that software breakthroughs could cool the hardware supercycle faster than analysts currently project, despite a consensus price target of $466.75 implying significant upside from the current $339 level.
The bull case for Micron remains anchored in strong fundamentals. The company's entire 2026 HBM capacity is already sold out, insulating it from any immediate demand shock. Furthermore, Micron reported a 169% year-over-year increase in NAND revenues to $5 billion in its second-quarter fiscal 2026 results, driven by higher prices and market share gains in solid-state drives. Wall Street remains broadly positive, with J.P. Morgan maintaining a Buy rating and a $550 price target.
However, the TurboQuant news provides a concrete catalyst for the bear case. The algorithm's ability to compress memory overhead for AI inference could, in the long term, reduce the need for physical memory, impacting Micron's core business. Compounding these fears are broader macroeconomic pressures, including geopolitical instability in the Middle East, and signs of institutional profit-taking. Wealthcare Advisory Partners and Net Worth Advisory Group, for instance, recently reduced their Micron holdings by 13.6% and 71.2%, respectively.
The TurboQuant story is still in its early stages, and its true impact on memory demand remains to be seen. For now, investors will be closely watching Micron's stock for signs of support around the $330 level. Any further commentary from analysts on the real-world implications of Google's new algorithm will likely dictate the stock's trajectory for the remainder of the week. The sharp downturn in retail sentiment, with social media scores turning "very bearish," suggests that the fear trade may have further to run.
This article is for informational purposes only and does not constitute investment advice.