IBM is shifting its strategy to focus on providing hybrid AI operating infrastructure for enterprise clients. The company utilizes a four-part architecture comprising agents, data, automation, and hybrid infrastructure, with Red Hat serving as the core open-source foundation for integration.
Analysts suggest that IBM aims to capitalize on the commoditization of AI models by managing them through its abstraction control layer. The company seeks to position itself as a provider of operational infrastructure rather than a proprietary model developer, targeting the integration of agents from various vendors.
Marvell Technology Inc. has set a revenue target of $70 billion to $90 billion for fiscal year 2031. The company, which produces custom AI processors and optical networking hardware, previously announced a partnership with Nvidia for NVLink Fusion technology.
The industry is experiencing a shift in AI workloads from training to inference. Analysts note that while training previously accounted for 90% of workloads, the ratio is moving toward 10% training and 90% inference as companies prioritize revenue generation through AI applications.
