Western Digital Corporation

(NASDAQ Global Select)
Back to all articles
Slightly Bullish +20

Semiconductor stocks sink as Google’s new AI tech threatens memory demand - Micron drops 4.82%, Sandisk tanks 8.35%

AI-generated analysis

📉 Semiconductor stocks slumped Thursday after Alphabet highlighted Google's TurboQuant algorithm aimed at optimizing storage efficiency for AI.

💾 Google states its new research can reduce memory capacity requirements for large language models by a factor of six, lowering training costs.

🤖 Investor anxiety stems from the potential that this technology could dampen aggressive demand for memory from hyperscale data centers.

📊 Shares of Micron Technology fell 4.82%, Western Digital lost 4.93%, Sandisk dropped 8.35%, and Nvidia slipped 2.60%.

💸 Hyperscalers like Amazon and Google are currently projected to allocate $650 billion this year toward data center construction purchases.

🔍 Morgan Stanley analyst Shawn Kim suggests the technology should be viewed positively as it addresses a vital technical bottleneck in AI inference.

⚖️ JPMorgan analysts note no immediate threat to overall memory consumption, though some investors may use the news to secure profits.

📉 South Korean market leaders Samsung Electronics and SK Hynix both saw their shares fall by at least 6% in Seoul trading.

🧠 Alphabet Inc.'s Google parent stock dropped 2.19% alongside other major tech giants including Microsoft, Meta, and Amazon.

⚡ TurboQuant is considered advantageous for hyperscalers due to improved return on investment and lower cost per token for AI queries.

Bullish Signals
  • Morgan Stanley analyst Shawn Kim highlighted that Google's TurboQuant algorithm addresses a vital technical bottleneck, viewing the broader industry impact as positive.
  • Analysts suggest that improved memory efficiency can enhance return on investment for hyperscalers, potentially driving higher product adoption demand and cost savings per token over the long term.
  • The technology could mitigate current supply shortages in AI hardware, which may help stabilize market pricing rather than causing a severe downturn.
  • While some investors might secure profits short-term, JPMorgan analysts maintained there is no immediate threat to overall memory consumption from this development.
Risk Factors
  • Semiconductor stocks sank following Google's announcement of the TurboQuant algorithm, which can reduce memory capacity requirements for large language models by a factor of six.
  • Investors are concerned that this technological advancement could dampen aggressive demand for memory from hyperscale data centers, potentially impacting component pricing for mobile devices and electronics.
  • Shares of chipmakers Micron Technology Inc. tumbled 4.82%, Western Digital Corp. lost 4.93%, Sandisk Corp. dropped 8.35%, Nvidia slipped 2.60%, and Advanced Micro Devices slid 6.16% on Thursday.
  • South Korean memory market leaders Samsung Electronics Co. and SK Hynix Inc. both saw their shares fall by at least 6% in Seoul trading despite the sector's earlier rally driven by AI investment.
  • While Morgan Stanley analyst Shawn Kim suggested the research addresses a technical bottleneck, investors remain pessimistic about the immediate impact on memory consumption valuations built during recent supply deficits.
Full Analysis
Shares of major semiconductor and memory companies fell sharply on Thursday after Alphabet Inc.'s Google highlighted a research breakthrough involving its TurboQuant algorithm, which reportedly reduces the memory capacity needed to run large language models by a factor of six. The technology, originally published last year, gained renewed attention on social media platform X, sparking investor concern that it could dampen demand for high-performance memory in hyperscale data centers and mobile devices. Following the news, Micron Technology Inc. shares tumbled 4.82%, Western Digital Corp. dropped 4.93%, and Sandisk Corp. plunged 8.35%. Other major chipmakers also saw significant declines, with Advanced Micro Devices sliding 6.16% and Nvidia slipping 2.60%, while Alphabet's parent company shares fell 2.19%. Analysts offered mixed interpretations of the market reaction. Morgan Stanley’s Shawn Kim argued that the broader industry impact should be viewed positively as it addresses a vital technical bottleneck in the inference phase of AI models, potentially lowering the cost to serve each query and enhancing return on investment for hyperscalers. He suggested that over the long term, lower costs per token could drive higher product adoption. However, other investors worried that such efficiency improvements might mitigate the aggressive demand previously driving supply deficits and surging chip valuations. JPMorgan analysts maintained that while some profit-taking was likely, there is no immediate threat to overall memory consumption from this single development. The stock market reaction came as hyperscalers led by Amazon.com Inc. and Google are projected to allocate approximately $650 billion this year toward data center construction, involving massive purchases of Nvidia Corp.'s AI accelerators and specialized memory units. Despite the rally memory manufacturers had experienced earlier due to intense investment in AI hardware, South Korean market leaders Samsung Electronics Co. and SK Hynix Inc. saw their shares fall by at least 6% in Seoul trading. Additionally, other tech giants faced declines, with Meta Platforms plunging 6.80%, Microsoft falling 1.02%, and Amazon declining 0.64%.