Photo Credit: StockSnap from Pixabay

Last week Intel (Nasdaq: INTC) reported its third quarter results – the first one since the U.S. Government became its biggest shareholder. The company returned to profitability and the strong results helped drive the stock 6% up in the after-hours trading session.

Intel’s Financials

For the quarter, revenues grew 3% over the year to $13.65 billion, ahead of the market’s estimates of $13.14 billion. It ended the quarter with an adjusted EPS of $0.23. Intel recorded a per-share loss of 37 cents in the quarter to account for shares in escrow that will be released to the U.S. government as part of its investment.

By segment, Intel’s Client Computing Group, which is primarily comprised of sales of central processors for PCs, saw revenues grow 3% to $8.5 billion. Revenue in the Data Center Group, which includes some AI chips, fell 1% to $4.1 billion. Foundry revenues were down 2% to $4.2 billion. Other revenues grew 3% to $1.0 billion.

For the current quarter, Intel expects revenues of $12.8-$13.8 billion, with the mid-point ahead of the market estimates of $13.37 billion. It expects to earn $0.08 per share for the quarter in line with the market estimates.

Intel’s AI Focus

These have been interesting few months for Intel. Besides the U.S. government announcing its $8.9 billion investment and the $5 billion that it received from Nvidia, Softbank invested $2 billion and it sold its stake in chipmaker Altera for $5.2 billion.

Intel is focused on leveraging its latest tie-up with Nvidia to accelerate its AI push. Its core x86 franchise is playing a critical role in its AI strategy. x86 is well-positioned to power the hybrid compute environment that AI workloads demand. The collaboration with Nvidia is helping Intel create a new class of products spanning multiple generations that accelerate the adoption of AI for the hyperscale, enterprise, and consumer markets. By integrating architectures through Nvidia NVLink, Intel will combine Intel CPU and x86 leadership with Nvidia’s AI and computing strengths to offer solutions that can deliver better customer experiences.

Intel realizes the need for a focused effort on the initiative, and recently created a central engineering group that will unify its horizontal engineering functions to drive leverage across foundational IP development, test chip design, EDA tools, and design platforms. The group will also lead the build-out of a new ASIC and design service business to deliver purpose-built silicon for external customers. This will extend the reach of core x86 IP but also leverage design strengths to deliver solutions from general-purpose to fixed-function computing.

Earlier this month, Intel announced that its fabrication plant in Chandler, Arizona, known as Fab 52, is now operational. Codenamed Panther Lake and Clearwater Forest, these SKUs will begin shipping late this year. If successful, these new CPUs will put Intel’s chipmaking capabilities in direct competition with TSMC. Intel is yet to disclose the expected yields from Fab 52. Panther Lake and Clearwater Forest are expected to also help Intel solidify its position in the notebook segment across both consumer and enterprise. Intel will still need to secure a large customer for its foundry chips to accelerate margins.

For its AI accelerator strategy, Intel is focused on developing compute platforms for emerging inference workloads driven by agentic AI and physical AI. In the shorter term, it will deliver AI capabilities through Xeon, AI PCs, ARC GPUs, and its open software stack. In the longer run, it plans to launch generations of inference-optimized GPUs on an annual cadence that will offer enhanced memory and bandwidth to meet enterprise needs.

Intel’s stock has climbed over 90% since the Trump government announced its plans earlier this summer. Its stock is trading at $40.16 with a market cap of $191.40 billion. It hit a 52-week high of $42.47 earlier this week. The stock has climbed from the 52-week low of $17.67 that it was trading at in April this year.

More By This Author: