Advanced Micro Devices (AMD) shares climbed 3.5% in Thursday trading, extending a recent rally as Wall Street analysts lifted price targets and the company's artificial intelligence narrative broadened beyond traditional GPU markets.
The move higher was supported by reports that Samsung could partner with AMD on chip production, offering an alternative to Taiwan Semiconductor Manufacturing Co. (TSMC) as its capacity remains constrained. Such a deal would provide AMD with additional manufacturing flexibility for its AI accelerators, a key concern for investors tracking supply chain risks.
Analysts Raise Targets on Expanding AI Opportunity
Several major financial institutions have boosted their AMD price targets in recent weeks, reflecting growing conviction in the company's positioning across AI infrastructure markets.
Bernstein increased its target to $600 from $525, maintaining an Outperform rating. The firm argued that the next phase of artificial intelligence—particularly agentic AI systems that can autonomously plan, search, and execute tasks—could substantially expand demand for server central processing units (CPUs). Bernstein now estimates the global server CPU market could reach $223 billion by 2030, up sharply from its prior forecast of $137 billion, describing the shift as a "CPU renaissance."
Baird raised its target to $625 from $300, citing stronger AI server positioning and potential gains in the data-center chip market. Citi upgraded AMD to Buy from Neutral, lifting its target to $575 from $460, and noted that investors may be underestimating the company's graphics processor opportunity and its potential to benefit from AI spending by Meta Platforms. Bank of America increased its target to $560 from $500, identifying AMD as a preferred CPU investment as AI inference and server processor demand expands. Barclays raised its target to $500, while Goldman Sachs upgraded the shares to Buy with a $450 price objective.
Agentic AI Strengthens AMD's Server CPU Outlook
Analysts increasingly believe agentic AI applications could materially alter the economics of AI infrastructure. Unlike traditional AI workloads focused on model training and inference, agentic systems can plan tasks, search databases, coordinate workflows, and execute multiple actions autonomously.
Bernstein argued that such workloads still require graphics processors but also depend heavily on CPUs for orchestration, memory management, and general-purpose computing. The firm believes CPU requirements in AI systems could increase significantly, potentially moving from earlier assumptions of one CPU supporting four to eight GPUs toward ratios approaching one-to-one in some inference-intensive applications. Because AMD already holds a strong competitive position in server CPUs through its EPYC line, analysts increasingly see its processor business as central to the company's AI investment thesis.
Manufacturing and Partnerships Broaden Growth Story
AMD's latest initiatives extend beyond conventional AI infrastructure. The company partnered with quantum technology firm EigenQ, which is pursuing a $3 billion special-purpose acquisition company transaction. The collaboration will focus on post-quantum cryptography, quantum-safe communications, and sensing technologies aimed at defense, industrial, and strategic applications.
AMD has also been actively expanding its AI footprint through investments and partnerships. Recently, the company entered a multi-year agreement with Rackspace Technology for a phased deployment of 30 megawatts of AMD-powered computing infrastructure between 2026 and 2028. The deployment will utilize Instinct GPUs and EPYC CPUs to support regulated enterprise and healthcare AI applications.
For broader context on AI market volatility, see our coverage of the AI Stock Rout Deepens as Iran Talks Ease Oil, Gold; Debt-Fueled Spending Worries Mount. Meanwhile, the Salesforce Hits 52-Week Low as Record 14-Day Losing Streak Deepens AI Fears highlights contrasting sentiment in the AI sector.
This article is for informational purposes only and does not constitute financial advice.
