Memory panic amid TurboQuant is a buying opportunity in these stocks: BofA

Investing.com — A sharp selloff in memory names following the debut of Google’s TurboQuant technology is creating an opportunity for investors, according to Bank of America.

Analyst Vivek Arya wrote in a note Friday that fears around high-bandwidth memory demand are overstated, noting that “AI capex remains the ultimate proof point of AI spend/demand, not efficiency measures.”

The concern stems from Google’s TurboQuant, which “effectively reduces KV cache memory footprint…by up to 6x without sacrificing accuracy.”

Shares of Micron Technology fell 7% Thursday, while the broader SOX dropped nearly 5%. But BofA argued that “compression techniques such as TurboQuant are not new,” pointing out that Google has published similar approaches for 18 months and simultaneously raised its 2026 capital-expenditure outlook by 100% to about $180 billion.

BofA said Google’s upcoming TPU v7 features “6x more HBM at 192GB vs. its prior TPUv6…at just 32GB,” while forthcoming chips from Nvidia, AMD and Meta also incorporate higher memory across generations.

“We continue to believe demand for AI memory remains strong,” the firm wrote, adding that efficiency gains are more likely to support larger model sizes than cut memory needs.

BofA also cited the DeepSeek episode in early 2025, when efficiency fears proved temporary, and China’s AI capex still rose 66% in 2025.

Preferred areas of exposure include AI compute, semicap equipment, networking and memory, with highlighted names such as Nvidia, Broadcom, Advanced Micro Devices, Lam Research, Applied Materials, Marvell Technology and Credo Technology.

Related articles

Memory panic amid TurboQuant is a buying opportunity in these stocks: BofA

5 reasons why Jefferies thinks Meta’s pullback is a buying opportunity

Wolfe Research outlines eight risks that could spark stock declines in 2026