From Dominance to Decline - What Could Undermine Nvidia's Competitive Edge?
$NVIDIA(NVDA.US$'s stock fell by 7% to $103.73, significantly outpacing the $Nasdaq Composite Index(.IXIC.US$'s 1.3% decline, marking its lowest closing share price since May 23. The stock is now down 26% from its June peak of over $140 per share.
Apple to Train AI Models with Google Chips Instead of Nvidia
Apple announced that its AI models, part of its Apple Intelligence system, were pretrained using Google's Tensor Processing Units (TPUs), indicating a shift among major tech companies away from Nvidia's GPUs for AI training. Detailed in a newly published technical paper, Apple's choice highlights the increasing competition in the AI chip market, dominated by Nvidia's high-demand, expensive GPUs, which are used by companies like OpenAI, Microsoft, and others.
Apple's use of Google's TPUs, rented from cloud providers, allows efficient and scalable training of various AI models, including on-device and server models. The move comes as Apple rolls out a preview of Apple Intelligence, featuring enhancements like a revamped Siri, improved natural language processing, and AI-generated summaries.
Despite being a latecomer to the AI race compared to peers who adopted generative AI post-ChatGPT's 2022 launch, Apple plans to introduce more AI-driven features over the next year, including image and emoji generation, and a more capable Siri. The paper also details the use of advanced TPU v5p and v4 chips for training different models.
Google's TPUs, cost-effective and designed for AI, have been available since 2017, though Google remains a significant Nvidia customer, utilizing both Nvidia GPUs and its TPUs for AI training and offering Nvidia's technology via its cloud services. This is Apple's second technical paper on its AI system, following a general version published in June.
Other Disturbing Factors
• A Sector Rotation
This drop is attributed to a combination of factors rather than a single event. After substantial gains, it is common for stocks to decline as investors take profits and adjust portfolios. Despite the recent drop, Nvidia's stock is still up 110% year-to-date and 610% since the end of 2022. Additionally, a slump in semiconductor stocks, influenced by former President Donald Trump's comments on trade and defense policies in East Asia, has affected Nvidia. The broader market's shift from big tech companies to smaller firms in anticipation of interest rate cuts also contributed to the decline. Other major tech companies like $Apple(AAPL.US$, $Microsoft(MSFT.US$, and $Alphabet-A(GOOGL.US$ have also experienced similar declines.
• Management Sell-Off
Nvidia CEO Jensen Huang has sold over 3 million shares since mid-June at an average price of about $126 per share, which can sometimes unsettle investors, especially given that the stock has more than doubled year-to-date. However, these sales were made through a prearranged trading plan known as Rule 10b5-1, allowing insiders to sell stock at scheduled times. Despite this, investors may view Huang's sales as a signal to reduce their own positions. Nvidia's fiscal second-quarter results are expected on August 28, and investors are already closely monitoring earnings reports in the sector.
Source: Forbes, Yahoo Finance, Market Watch, WEVOLVER, CNBC
Disclaimer: Moomoo Technologies Inc. is providing this content for information and educational use only.
Read more
Comment
Sign in to post a comment
Moomoo News Official Account
How do analysts evaluate the market?
17KFollowers
46Following
36KVisitors
Follow