Intel made significant announcements Monday, but investor sentiment remained mixed. The semiconductor manufacturer saw its shares decline roughly 5% in premarket activity to approximately $109, retreating from its 52-week peak of $132.75 reached in May.
The chipmaker revealed its latest AI initiative: a GPU named Crescent Island, engineered primarily for inference operations rather than the more contested training segment. Initial shipments are projected to begin by late 2026.
Kevork Kechichian, leading Intel’s data-center division, explained to the Financial Times that the company made a strategic decision to avoid direct competition with NVIDIA in the AI training arena, citing lessons from its Gaudi chip initiative’s underwhelming performance.
The fundamental principle driving Crescent Island is affordability. Rather than utilizing high-bandwidth memory (HBM) and liquid cooling systems standard in NVIDIA and AMD products, Intel’s chip incorporates LPDDR5 memory paired with conventional air cooling.
This approach establishes it as a more economical alternative for AI customers seeking capable inference capabilities without requiring flagship GPU performance.
Kechichian additionally mentioned Intel is evaluating potential sales of specific chip variants in China, provided they comply with U.S. export regulations — indicating substantial demand in that price segment.
Under CEO Lip-Bu Tan’s leadership, Intel intends to produce Crescent Island through its proprietary foundries, potentially reducing expenses compared to competitors relying on TSMC manufacturing.
This internal production strategy represents a cornerstone of Intel’s comprehensive restructuring effort, which has generated largely favorable investor responses since new management assumed control.
The timing proved challenging. Coinciding with Intel’s AI hardware presentation, NVIDIA unveiled the RTX Spark Superchip — a processor engineered to compete head-to-head in the PC processor arena where Intel has historically dominated.
NVIDIA CEO Jensen Huang described the chip, developed in partnership with MediaTek and compatible with Windows, as designed to enable AI agents capable of executing cross-application tasks with limited user involvement. NVIDIA and Microsoft allegedly collaborated on the platform for three years.
This strategic move applies pressure on Intel simultaneously across AI infrastructure and its traditional PC business segments.
Institutionally, Intel continues attracting investment interest. Several firms recently increased their positions, with institutional investors controlling 64.53% of outstanding shares.
The company’s latest quarterly results substantially exceeded projections — delivering $0.29 EPS versus consensus estimates of $0.01, while revenue of $13.58 billion surpassed forecasts of $12.32 billion.
For upcoming quarters, Wall Street anticipates Intel will post Q2 EPS around $0.19 and full-year EPS of approximately $0.63. The company’s next earnings announcement is projected for roughly July 23, 2026.
Citigroup elevated its price objective to $130 while Benchmark increased its target to $140, both adjustments occurring in May. However, the consensus analyst target remains at $77.65, indicating skepticism persists regarding the stock’s recent gains.
The post Intel (INTC) Shares Slide 5% as NVIDIA Invades Core PC Market Territory appeared first on Blockonomi.
