Nvidia's 66% Margin Tops SpaceX's 830% Revenue Growth

Key Takeaways

• Nvidia's 66% operating margin far exceeds SpaceX's 5.7% despite SpaceX's 830% revenue growth in Q3
• Amazon plans to spend $220 billion on capital expenditures in 2026 while reporting a $42.4 billion free cash flow loss
• Anthropic signed $224 billion in cloud deals with Google ($114.5B) and Amazon ($110B) over the next decade
• Intel's Data Center and AI unit earned $5.32 billion in operating income, surpassing its PC unit's $3 billion
• Samaya AI's Sterling model beats Claude Fable 5.1 on finance benchmarks at 11x lower cost
• Alphabet shares fell 13% as AI spending and weak ad sales shook markets
• A circuit board shortage is hitting PC makers as factories prioritize AI server production

SpaceX and Nvidia compete as top AI stocks

SpaceX and Nvidia are both leading AI investments but in different ways. SpaceX's AI division Super Intelligence saw revenue grow 830% in Q3, while Nvidia grew 106%. Nvidia has a 66% operating margin compared to SpaceX's 5.7%. Nvidia trades about three times cheaper than SpaceX, making it the safer AI stock pick according to analysts.

Amazon spends heavily on AI while losing cash

Amazon spent $77.3 billion on property and equipment in Q3, more than double what it spent a year earlier. The company reported a free cash flow loss of $42.4 billion over the past 12 months. AWS revenue reached $46.0 billion with $17.8 billion in operating profit. Amazon plans to spend about $220 billion on capital expenditures in 2026 and raised $48.1 billion in debt.

Anthropic's $224 billion cloud deals reveal AI winners

Anthropic plans to spend about $224 billion on cloud services with Amazon and Alphabet over the next decade. The AI lab will pay about $114.5 billion to Google Cloud through 2033 and an additional $110 billion to Amazon through 2036. Anthropic does not have its own computing infrastructure and relies on these partners. Alphabet's custom Tensor Processing Units are also part of the deal.

Amazon pushes Alexa against Meta's Muse shopping tool

Amazon CEO Andy Jassy promoted Alexa as a shopping assistant while criticizing Meta's Muse. Over 400 million customers used Alexa for shopping in the past year, with interactions up 570%. US Alexa shoppers spend 50% more per order than non-users. Meta says more than 10 million people in the US and Canada set up Muse, and over a quarter of users shopped with an average purchase over $200.

Amazon CEO says AI shopping agents still have problems

Amazon CEO Andy Jassy said AI shopping agents have at least two problems right now. He noted that few customers are shopping this way and the experience is still not better than going directly to a retailer. Amazon blocked Meta's Muse from its site in September over data privacy concerns. Amazon's ad business earned $19.6 billion in Q3, up 12% year over year.

Intel's AI unit earns $3B, more than PCs, as sales jump 78%

Intel reported that its Data Center and AI unit made $5.32 billion in operating income in the third quarter, a 57% increase from a year earlier. This was more than the $3 billion earned by Intel's PC unit. The Data Center and AI unit made up about 57% of Intel Products' total operating income. Shortages of some raw materials and a factory shutdown in Israel limited the unit's performance.

Samaya AI releases Sterling, a finance research model that beats Claude Fable 5.1

Samaya AI launched Sterling, a 30-billion-parameter model with 3 billion active parameters designed for finance research. Sterling scores 58.9% on the FrontierFinance benchmark, beating Claude Fable 5.1 which scores 55.9%. It costs $0.61 per question compared to Fable 5.1's $6.89, an 11x reduction in cost. The model was trained using four stages including synthetic data training and GRPO methods.

Musk and Zuckerberg echo Trump's push to rebrand AI as super intelligence

Meta's Mark Zuckerberg and Elon Musk used the term super intelligence instead of artificial intelligence on their third-quarter earnings calls, following President Trump's push. Trump ordered the term super intelligence to replace artificial intelligence in government communications last month. However, the CEOs of Microsoft, Amazon, and Alphabet continued to use the term AI on their earnings calls.

AI security startup AvalonX emerges from stealth with $10 million in funding

AvalonX launched after 18 months in stealth mode with $10 million in funding from Ridown Capital. Its Artemis platform uses a 400-billion-parameter security AI called Aether trained on nine years of attack and defense data. The platform offers continuous penetration testing as a monthly subscription for small and medium-sized businesses. It identifies intrusion paths before cybercriminals can exploit them.

Weaken to Strengthen: New self-distillation method improves reasoning models

Researchers introduced Weakening-Induced On-Policy Self-Distillation, called WISD, to improve large reasoning models without needing privileged information. The method builds a weaker version of the original model through quantization or pruning and uses the contrast as a training signal. Pruning raised Qwen3.5-4B's accuracy from 57.7% to 70.5% on HMMT 2025. Quantization raised Qwen3.6-35B-A3B's accuracy from 68.5% to 74.6% on the same benchmarks.

Factory acquires Typedef to add data context to autonomous coding

Factory acquired Typedef to help engineers understand data relationships when Droids build software. Typedef creates a context layer that shows how code, pipelines, and dashboards connect. The founders Yoni Michael and Kostas Pardalis bring deep experience from companies like Coolan, Blendo, Salesforce, and Tecton. Factory, founded in 2023 in San Francisco, aims to combine its Droid-based platform with Typedef's data lineage tools.

Alphabet shares plunge as AI spending and weak ad sales shake markets

Alphabet shares fell about 13% on October 29 after ad revenue missed expectations and AI spending grew. The drop could have wiped out roughly $500 billion in market value and dragged the Nasdaq lower. AI Mode searches brought in less money than traditional searches, adding investor concern. U.S. GDP grew at a 3.4% annualized rate in the third quarter, while core PCE inflation rose at a 2.5% pace.

AI demand triggers shortage of circuit boards for PCs and medical gear

A shortage of printed circuit boards is hitting PC makers like HP and Dell because factories focus on AI server boards. High-end AI server boards use more materials and factory time, pushing up laminate costs by at least 12 times since July. Amazon bought a stake in Taiwanese board maker Gold Circuit Electronics to secure supply. Canon expects to lose about ¥90 billion in medical equipment sales due to board shortages that began in August.

Sources

• SpaceX vs. Nvidia: The Best AI Stock to Buy Now
• Amazon Just Burned $42 Billion in Cash Chasing AI Growth. Is Wall Street Ignoring the Risk?
• Anthropic's $224 Billion Cloud Bill Reveals the Real Winner of the AI Boom
• Amazon wants shoppers to use Alexa, not Meta's Muse. Here's why the AI shopping battle is getting personal? Decoded here
• Amazon CEO Andy Jassy Says AI Shopping Agents Have at Least 2 Problems Right Now
• Intel's AI Unit Earns $3B, More Than PCs, as Sales Jump 78%
• Samaya AI Releases Sterling: A 3B-Active Finance Research Agent Model that Beats Claude Fable 5.1 on FrontierFinance at 11x Lower Cost
• Musk and Zuckerberg echo Trump's push to rebrand AI as 'super intelligence' on earnings calls. Others did not
• AI security startup AvalonX emerges from stealth with $10 million in funding from Ridown Capital
• Weaken to Strengthen: Weakening-Induced On-Policy Self-Distillation for Large Reasoning Models
• Factory Acquires Typedef: Bringing Data Context to the Software Factory
• What Alphabet's Steep Drop Means for Investors and the AI Trade
• The AI hardware crunch has reached the humble circuit board