In a transformative analysis released on April 20, 2026, Morgan Stanley has highlighted a significant shift in the landscape of artificial intelligence (AI) and its impact on semiconductor spending. As agentic AI—characterized by increasingly autonomous decision-making systems—gains prominence, it is expected to alter the traditional demand for computing resources, specifically transitioning from graphics processing units (GPUs) to central processing units (CPUs) and memory. This shift is poised to generate substantial growth in the data-center CPU market, potentially adding between $32.5 billion to $60 billion by 2030, in a sector already exceeding $100 billion.

The Autonomous AI Revolution

Agentic AI refers to systems capable of performing tasks with a degree of independence, evolving from traditional AI, which primarily relies on human inputs and supervision. The emergence of such autonomous systems requires a robust infrastructure capable of handling increased computational demands, thus spotlighting the vital roles played by CPUs and memory components.

The Shift from GPUs to CPUs

Historically, GPUs have dominated the AI landscape due to their parallel processing capabilities, which are well-suited for handling the vast amounts of data required for training complex models. However, Morgan Stanley’s analysis indicates that as AI systems advance toward autonomous functions, the reliance on CPUs and memory will intensify. This is because:

• CPU Performance: CPUs are essential for general-purpose computing tasks that support the orchestration and management of AI processes.
• Memory Bottlenecks: With increased AI autonomy, the demand for memory to support rapid data access and processing will surge.
• Complex Workflows: Autonomous AI systems will necessitate more intricate workflows that require the computational abilities of CPUs.

Market Implications

The anticipated growth in CPU and memory demand poses significant implications for the semiconductor market. Morgan Stanley’s assessment suggests that the transition will not only benefit CPU manufacturers but also bolster memory suppliers and chipmakers, leading to a diversified investment landscape. This evolving demand will encourage:

• Increased Investment: Stakeholders in the semiconductor industry may redirect their investments from solely GPU-focused projects to include CPUs and memory solutions.
• Manufacturing Innovations: As chipmakers adapt to meet the new demands, innovations in manufacturing processes may emerge to enhance CPU and memory capabilities.
• Broader Market Growth: The surge in demand for CPUs and memory could catalyze broader growth across the technology sector, impacting various industries reliant on AI.

The Data Center Expansion

With the projected increase in CPU market growth, data-center expansions are likely to accelerate. As businesses and organizations seek to accommodate the burgeoning computational needs of autonomous AI systems, the following trends may emerge:

• Expanded Infrastructure: Companies will invest in larger data centers equipped with advanced CPU and memory technologies to support AI workloads.
• Energy Efficiency Focus: As data centers expand, the need for energy-efficient solutions will become paramount, leading to innovations in cooling and power management.
• Hybrid Architectures: Future data centers may adopt hybrid architectures that leverage both CPU and GPU resources to optimize performance for diverse workloads.

Challenges Ahead

While the prospects for CPU and memory markets appear promising, several challenges could impede this transition:

• Supply Chain Constraints: The semiconductor industry has faced supply chain disruptions, which could affect the availability of CPUs and memory components.
• Competition for Resources: As demand increases, competition for critical materials required for chip manufacturing may intensify.
• Technological Barriers: The evolution of AI capabilities may outpace advancements in hardware, leading to potential bottlenecks.

Conclusion

The insights provided by Morgan Stanley mark a pivotal moment in the intersection of AI and semiconductor markets. As agentic AI systems become more autonomous, the demand for CPUs and memory is set to soar, fundamentally reshaping the investment landscape within the technology sector. Stakeholders must adapt to these changes, embracing the opportunities while navigating the challenges presented by this new era of artificial intelligence.

In summary, the transition from GPU to CPU-centric architectures represents not only a shift in technology but also a broader evolution in how AI will be integrated into various sectors. As organizations prepare for this emerging reality, the spotlight will undoubtedly shine on the semiconductor industry, which is poised to play a crucial role in the advancement of autonomous artificial intelligence.