Ken Griffin, founder of Citadel Advisors, made headlines in Q3 with a portfolio rebalancing that tells an interesting story about AI opportunities. While his hedge fund—historically the most successful by net gains since inception—divested 1.6 million Amazon shares, it simultaneously acquired 388,000 shares of Palantir Technologies. The timing is noteworthy: Palantir has delivered 1,030% returns since January 2024, significantly outpacing Nvidia’s 281% gain over the same period.
This move raises an important question: Is Griffin abandoning his confidence in one of the world’s largest companies, or simply rotating capital toward emerging opportunities? The answer reveals much about where sophisticated money is flowing in the AI era.
Amazon’s AI Transformation Is Working—But Is It Enough?
Amazon remains a formidable player across three high-growth industries: e-commerce, digital advertising, and cloud computing. The company has aggressively integrated generative AI throughout its operations with tangible results.
On the e-commerce front, Amazon deployed AI-powered tools for customer service, inventory optimization, and last-mile delivery efficiency. Its AI shopping assistant Rufus is projected to generate $10 billion in sales this year alone. In advertising, the third-largest ad-tech player created generative AI capabilities enabling brands to produce images, video, and audio at scale. Amazon Web Services (AWS), the market-leading public cloud provider, introduced custom AI chips for training and inference that offer cost advantages versus Nvidia GPUs, alongside new agentic AI tools automating software development and security operations.
The financial impact speaks for itself. Q3 revenue surged 13% to $180 billion, with advertising and cloud segments accelerating particularly quickly. Operating margin expanded 60 basis points (excluding one-time items), while operating income jumped 23% to $21.7 billion. Wall Street projects 18% annual earnings growth over the next three years, making the current 33x earnings multiple defensible.
So why would Ken Griffin trim Amazon? The most likely explanation is disciplined profit-taking. Notably, Amazon remains among Citadel Advisors’ ten largest positions—hardly the move of a skeptic.
Palantir’s Explosive Growth Masks a Dangerous Valuation Problem
Palantir Technologies operates in a different league entirely. The company’s data analytics and AI platforms serve public and private sector clients through proprietary ontology-based software—essentially decisioning frameworks continuously refined by machine learning models. Its applications range from supply chain optimization to battlefield analytics to financial fraud detection.
Industry recognition has been substantial. Forrester Research ranked Palantir as the market’s most capable AI/ML platform, ahead of Google Cloud, AWS, and Microsoft Azure. The company’s financial trajectory has matched this validation: Q3 revenue accelerated 63% to $1.1 billion for the ninth consecutive quarter, while non-GAAP earnings more than doubled to $0.21 per diluted share. Management attributed the surge directly to robust demand for its AI platform.
Here’s where the story gets complicated. Palantir trades at 119 times sales—the highest valuation in the entire S&P 500, dwarfing AppLovin at 45x. The math is brutal: Palantir’s stock price increased 11x since January 2024, but revenue expanded less than 2x. The stock traded at 18x sales in January 2024. This means almost all gains reflect multiple expansion rather than fundamental business improvement.
Put differently, Palantir could lose over 60% of its value and still command the most expensive valuation in the index. That gap cannot persist indefinitely.
Ken Griffin’s Position: Conviction or Calculation?
While Ken Griffin’s acquisition of Palantir shares generated headlines, context matters. The position does not rank among Citadel Advisors’ top 300 holdings, suggesting measured conviction rather than bold conviction.
The takeaway for investors: Sophisticated capital is selectively rotating within the AI theme. Amazon represents the proven, profitable AI narrative already priced in. Palantir represents explosive growth hijacked by speculative excess. When multiple compression inevitably arrives—and history suggests it will—Palantir faces material downside risk despite genuine business quality.
This page may contain third-party content, which is provided for information purposes only (not representations/warranties) and should not be considered as an endorsement of its views by Gate, nor as financial or professional advice. See Disclaimer for details.
Ken Griffin's Portfolio Shift: Why One Hedge Fund Legend Abandoned Amazon for a 1,030% AI Winner
The Big Move That Caught Wall Street’s Attention
Ken Griffin, founder of Citadel Advisors, made headlines in Q3 with a portfolio rebalancing that tells an interesting story about AI opportunities. While his hedge fund—historically the most successful by net gains since inception—divested 1.6 million Amazon shares, it simultaneously acquired 388,000 shares of Palantir Technologies. The timing is noteworthy: Palantir has delivered 1,030% returns since January 2024, significantly outpacing Nvidia’s 281% gain over the same period.
This move raises an important question: Is Griffin abandoning his confidence in one of the world’s largest companies, or simply rotating capital toward emerging opportunities? The answer reveals much about where sophisticated money is flowing in the AI era.
Amazon’s AI Transformation Is Working—But Is It Enough?
Amazon remains a formidable player across three high-growth industries: e-commerce, digital advertising, and cloud computing. The company has aggressively integrated generative AI throughout its operations with tangible results.
On the e-commerce front, Amazon deployed AI-powered tools for customer service, inventory optimization, and last-mile delivery efficiency. Its AI shopping assistant Rufus is projected to generate $10 billion in sales this year alone. In advertising, the third-largest ad-tech player created generative AI capabilities enabling brands to produce images, video, and audio at scale. Amazon Web Services (AWS), the market-leading public cloud provider, introduced custom AI chips for training and inference that offer cost advantages versus Nvidia GPUs, alongside new agentic AI tools automating software development and security operations.
The financial impact speaks for itself. Q3 revenue surged 13% to $180 billion, with advertising and cloud segments accelerating particularly quickly. Operating margin expanded 60 basis points (excluding one-time items), while operating income jumped 23% to $21.7 billion. Wall Street projects 18% annual earnings growth over the next three years, making the current 33x earnings multiple defensible.
So why would Ken Griffin trim Amazon? The most likely explanation is disciplined profit-taking. Notably, Amazon remains among Citadel Advisors’ ten largest positions—hardly the move of a skeptic.
Palantir’s Explosive Growth Masks a Dangerous Valuation Problem
Palantir Technologies operates in a different league entirely. The company’s data analytics and AI platforms serve public and private sector clients through proprietary ontology-based software—essentially decisioning frameworks continuously refined by machine learning models. Its applications range from supply chain optimization to battlefield analytics to financial fraud detection.
Industry recognition has been substantial. Forrester Research ranked Palantir as the market’s most capable AI/ML platform, ahead of Google Cloud, AWS, and Microsoft Azure. The company’s financial trajectory has matched this validation: Q3 revenue accelerated 63% to $1.1 billion for the ninth consecutive quarter, while non-GAAP earnings more than doubled to $0.21 per diluted share. Management attributed the surge directly to robust demand for its AI platform.
Here’s where the story gets complicated. Palantir trades at 119 times sales—the highest valuation in the entire S&P 500, dwarfing AppLovin at 45x. The math is brutal: Palantir’s stock price increased 11x since January 2024, but revenue expanded less than 2x. The stock traded at 18x sales in January 2024. This means almost all gains reflect multiple expansion rather than fundamental business improvement.
Put differently, Palantir could lose over 60% of its value and still command the most expensive valuation in the index. That gap cannot persist indefinitely.
Ken Griffin’s Position: Conviction or Calculation?
While Ken Griffin’s acquisition of Palantir shares generated headlines, context matters. The position does not rank among Citadel Advisors’ top 300 holdings, suggesting measured conviction rather than bold conviction.
The takeaway for investors: Sophisticated capital is selectively rotating within the AI theme. Amazon represents the proven, profitable AI narrative already priced in. Palantir represents explosive growth hijacked by speculative excess. When multiple compression inevitably arrives—and history suggests it will—Palantir faces material downside risk despite genuine business quality.