Cathie Wood's ARK Sells AMD at $479.18, Buys Cerebras Systems
0h ago|5 min readStandard
Fazen Markets Editorial Desk
Collective editorial team · methodology
cathie-woodark-invest
Fazen Markets Editorial Desk
Collective editorial team · methodology
Trades XAUUSD on autopilot. Verified Myfxbook performance. Free forever.
Risk warning: CFDs are complex instruments and come with a high risk of losing money rapidly due to leverage. The majority of retail investor accounts lose money when trading CFDs. AiX is informational software — not investment advice. Past performance does not guarantee future results.
ARK Investment Management, led by Cathie Wood, sold its position in Advanced Micro Devices Inc. and allocated capital to private artificial intelligence hardware developer Cerebras Systems. The transaction was executed on August 25, 2026, as AMD shares traded within a daily range of $469.07 to $481.50. This portfolio adjustment reflects a continued strategic emphasis on disruptive innovation within the AI ecosystem, moving from a public semiconductor leader to a private company specializing in wafer-scale compute.
ARK's sale of a major semiconductor holding coincides with a period of exceptional performance for the sector. The trade occurs as the Philadelphia Semiconductor Index trades near all-time highs, fueled by relentless demand for AI-training and inference chips. This marks a notable shift in ARK's public market strategy, which has historically included significant allocations to GPU and CPU manufacturers like NVIDIA and AMD.
The macroeconomic backdrop features stable long-term Treasury yields and moderating inflation, conditions that typically support growth equity valuations. ARK's transaction signals a potential reassessment of risk and reward within the public versus private AI infrastructure market. The move to invest in Cerebras, a company developing an alternative to GPU clusters, indicates a belief that next-generation AI training may require novel architectures.
Cerebras Systems has gained attention for its Wafer Scale Engine, the largest computer chip ever built. The company competes directly with established players like AMD and NVIDIA by offering a different technical approach to accelerating large language model training. ARK's investment thesis has long focused on identifying winners in technological paradigm shifts, often before they are widely recognized by the public market.
The last major portfolio shift of this nature occurred in the third quarter of 2025, when ARK reduced its Tesla position by 15% to increase exposure to autonomous driving and robotics startups. That trade preceded a period of consolidation for Tesla's stock and significant funding rounds for several private companies in the ARK portfolio.
At the time of the sale, AMD stock traded at $479.18, representing a daily gain of 1.25%. The share price reached an intraday high of $481.50 while maintaining support above $469.07. AMD's market capitalization stands at approximately $773 billion, ranking it among the top five semiconductor companies globally by valuation.
AMD's performance year-to-date has significantly outpaced the broader technology sector, with gains exceeding 45% compared to the Nasdaq 100's advance of 22%. The stock's 30-day average trading volume has exceeded 48 million shares, indicating substantial institutional interest. AMD's price-to-earnings ratio of 58 reflects premium valuation metrics common among high-growth AI infrastructure companies.
Cerebras Systems remains privately held, with its latest funding round in April 2026 valuing the company at approximately $8 billion. This valuation represents a substantial increase from its $4 billion valuation in late 2024. The company has raised over $1.8 billion in total funding from investors including Altimeter Capital and Benchmark.
The AI hardware market is projected to reach $250 billion annually by 2028, according to industry analysts. Cerebras claims its systems can train AI models up to 90% faster than comparable GPU clusters while consuming less power. This performance advantage forms the core investment thesis for companies betting on alternative AI architectures.
| Metric | AMD | Cerebras (Private) |
|---|---|---|
| Valuation | $773B | ~$8B |
| YTD Performance | +45%+ | N/A |
| P/E Ratio | 58 | N/A |
ARK's divestment from AMD creates immediate selling pressure of approximately $85-120 million, based on the fund's typical position sizing. This supply could temporarily weigh on AMD's momentum, particularly if other growth-focused funds follow suit. The transaction may benefit competing AI hardware developers like NVIDIA, which could see increased institutional interest as the purest public market play on AI acceleration.
Semiconductor equipment manufacturers including Applied Materials and Lam Research may experience secondary effects, as reduced capital allocation to public semiconductor companies could signal caution about future capital expenditure cycles. The VanEck Semiconductor ETF experienced increased volume following the news, suggesting traders are monitoring sector rotation patterns.
A counter-argument suggests that ARK's move reflects specific portfolio management needs rather than a fundamental view on AMD's prospects. The fund may be rebalancing to maintain diversification standards or meeting redemption requests, rather than making a strategic judgment about AMD's competitive position. AMD continues to gain market share in data center CPUs and GPUs, with recent quarterly revenue exceeding analyst expectations by 8%.
Flow data indicates institutional buyers absorbed ARK's AMD position quickly, with block trades executing near the day's high. Hedge funds and quantitative strategies appear to be adding AMD exposure despite ARK's sale, suggesting divergent views on the stock's valuation. Options activity shows increased demand for January 2027 calls at the $500 strike price.
Market participants should monitor Cerebras Systems' filing activity with the Securities and Exchange Commission for signals of an impending initial public offering. The company has not publicly disclosed IPO plans, but a successful funding round at an $8 billion valuation typically precedes a public listing within 12-18 months.
AMD's next earnings release on October 22, 2026, will provide crucial data on whether AI-related revenue can justify current valuations. Analysts project data center revenue growth of 35% year-over-year, with particular strength in MI300 series accelerator sales. Guidance for 2027 will be scrutinized for any moderation in AI spending forecasts.
The semiconductor sector faces a key test on September 15, 2026, when the Semiconductor Industry Association releases global sales data for August. Previous months have shown 18% year-over-year growth, with any deviation from this trend likely affecting sector valuations. Taiwan Semiconductor Manufacturing Company's monthly sales report on October 10 will provide additional insight into foundry demand trends.
Technical analysts identify $485 as critical resistance for AMD shares, a level that has contained advances twice in the past three months. Support appears firm at $465, representing the 50-day moving average. A sustained break above $485 could trigger renewed momentum buying, while a drop below $465 might indicate a broader sector rotation.
ARK's sale represents a single institutional decision that may not reflect broader market sentiment. AMD shares traded positively following the news, reaching $481.50, suggesting other buyers were willing to absorb the shares without price concession. Long-term price movement will depend more on AMD's execution in AI markets than on a single fund's positioning. The stock remains up significantly year-to-date despite this selling pressure.
Cerebras competes by offering wafer-scale solutions that differ fundamentally from the discrete GPU approach favored by established players. Their Wafer Scale Engine integrates processor cores on a single silicon wafer rather than using multiple interconnected chips. This architecture aims to reduce communication latency and simplify programming models for massive AI training workloads, though it faces challenges in manufacturing yield and ecosystem support.
Wood's strategy focuses on identifying enabling technologies before they achieve mainstream adoption, often emphasizing private companies with disruptive potential. The approach typically involves early investment in innovation, willingness to accept higher volatility, and concentration in themes like artificial intelligence, robotics, and genomics. This strategy has produced both spectacular successes and significant drawdowns depending on market conditions and technological adoption curves.
ARK's trade exchanges public semiconductor exposure for private AI infrastructure potential.
Disclaimer: This article is for informational purposes only and does not constitute investment advice. CFD trading carries high risk of capital loss.
AiX is our free MetaTrader 4 Expert Advisor. Verified Myfxbook performance. No subscription. No fees. XAUUSD breakout engine.
Trade 800+ global stocks & ETFs
Start TradingSponsored
Open a demo account in 30 seconds. No deposit required.
CFDs are complex instruments and come with a high risk of losing money rapidly due to leverage. You should consider whether you understand how CFDs work and whether you can afford to take the high risk of losing your money.