In late January, a leading Wall Street analyst downgraded Intel Corporation to a sell rating, citing a loss of momentum in the company’s long-promised recovery. Fourth-quarter revenue fell 4% year-over-year to $13.67 billion, while gross margin contracted by 180 basis points to 49.3%. Management guided first-quarter revenue below consensus estimates, indicating that the enterprise and PC markets remain soft. The analyst noted that Intel’s valuation has expanded sharply over the past six months, with much of the optimism for a successful turnaround already priced in despite weakening fundamentals.

At the Cisco AI Summit, Intel CEO Lip-Bu Tan confirmed that Intel will enter the graphics processing unit market, traditionally dominated by a handful of rivals. The GPU project will be led by Kevork Kechichian, executive vice president of the Data Center Group, and Eric Demmers, hired in January after 13 years at Qualcomm. While still in early stages, the initiative aims to address growing demand for AI training and high-performance computing, with design work now underway and customer-driven requirements shaping the product roadmap.

Investor enthusiasm for Intel’s foundry segment helped fuel a 26% stock rally in January, driven by expectations for contract manufacturing growth and recent government subsidies totaling $10 billion. However, the business continues to lack a marquee external customer. Management has highlighted ongoing discussions with prospective partners, but no firm commitments have been announced. Production capacity is being repurposed for server CPU lines, which may delay dedicated foundry output until late 2026, according to internal planning documents.

Intel unveiled its Xeon 600 family of processors in December, featuring up to 36 cores, on-chip AI acceleration engines, and upgraded connectivity for high-throughput data center workloads. Benchmarks published by early adopters show single-socket performance gains of up to 40% versus prior-generation models in machine-learning inference tasks. The company is targeting both traditional workstation customers and hyperscale cloud providers, positioning the new chips as a bridge between compute and specialized AI accelerators.