Intel announced its recovery strategy centered on next-generation data center processors.

Intel faces significant challenges in its data center business following lost server CPU market share to AMD, delayed product releases, and consecutive quarters of year-over-year revenue declines. The chipmaker has nevertheless outlined an ambitious product roadmap aimed at restoring growth across multiple markets, including its first-ever foray into GPUs for high-performance computing, artificial intelligence, and media processing. Industry analysts acknowledge Intel has a solid product strategy but question whether the company can execute effectively. Manoj Sukumaran, Omdia's principal analyst of data center IT, emphasized that "It's not just having the silicon. It takes a lot of effort and time to create the software ecosystem and the adoption among the developer community to come up with workloads and applications optimized for specific silicon features," adding that "the competition is not staying idle. They are also working hard to come up with the next generation of processors."

Intel's market position has eroded substantially. AMD's datacenter CPU share has doubled from 10% in late 2020 to 22% in mid-2022, while Intel's share dropped from 87% to 70% during the same period, according to Omdia's estimates. Arm CPU adoption in the data center has risen to 7%, with nearly all hyperscale cloud service providers using Arm CPUs that cost significantly less than x86 processors. Intel has also ceded the lucrative GPU market to Nvidia's dominance and entered the Data Processing Unit market late, after both Nvidia—through its Mellanox acquisition—and AMD—through its Pensando Systems acquisition—established footholds.

Under CEO Pat Gelsinger's leadership, Intel has announced four new Xeon processors through 2024, beginning with the next-generation "Sapphire Rapids," which the company says will deliver significant performance boosts for AI and HPC workloads. However, Sapphire Rapids has faced multiple delays and is now expected to ramp later in 2022 with limited shipments by year-end. In the GPU market, Intel unveiled plans for three data center GPUs: "Ponte Vecchio," currently being sampled by a limited number of customers with a 2022 release planned; "Rialto Bridge"; and "Falcon Shores," targeted for 2024. Jeff McVeigh, vice president and general manager of Intel's Super Compute Group, stated that "Ponte Vecchio really has been designed for this convergence between HPC and AI…. We feel it will compete very well from a standalone performance standpoint," and added: "We are new to this high-end GPU space and HPC and AI world, so we'll also make sure that it's delivering great TCO to our customers." Both the Sapphire Rapids Xeon and Ponte Vecchio GPU will power the Aurora exascale supercomputer at Argonne National Laboratory.

Intel's strategic approach extends beyond CPUs and GPUs. The Falcon Shores GPU, planned for 2024, will combine an x86 CPU and GPU into a single socket—a strategy mirroring Nvidia's Grace Hopper Superchip and AMD's planned approach. Kevin Krewell, principal analyst at Tirias Research, explained that "The goal is to bring the CPU and GPU closer together for accelerated compute…. By having them in the same package, you cut down on latency, and with shared memory, you get performance benefits." Additionally, Intel released its Data Center GPU Flex Series in August for media processing, AI visual inference, cloud gaming, and desktop virtualization. The company is entering the Data Processing Unit market with its IPU offering, collaborating with Google Cloud on an ASIC-based IPU code-named Mount Evans. Intel's portfolio is further bolstered by FPGAs acquired from Altera in 2015 and AI chips for training and inference acquired from Habana Labs in 2019. Krewell summarized Intel's approach: "Intel is basically going after the data center market with everything it's got. They've pretty much gone with a shotgun approach and offer every type of solution for the data center, so customers can find a solution that works for them."