AI Hardware Updates: October 2026

From major AI chip financing deals and semiconductor supply constraints to new smart home devices and XR glasses, October has brought several developments across the AI hardware industry.

Here's a roundup of the latest announcements, investments, and product launches.

AI chips and semiconductor news

1. Huawei expands its AI chip roadmap to challenge Nvidia

Huawei says demand for its AI computing systems already exceeds its production capacity in China. Its roadmap includes the Ascend 950DT processor and next-generation chips designed to help Chinese companies train and run AI models using domestic hardware.

The company is also developing systems that connect large numbers of processors, addressing communication bottlenecks that can slow down large AI workloads. Nvidia retains a major software advantage through its widely used CUDA platform, which remains a challenge for competing chipmakers. Read the full story.

2. OpenAI reportedly plans to buy Google's AI chips

OpenAI is reportedly planning to purchase Google's Tensor Processing Units (TPUs), adding another option alongside the Nvidia GPUs it uses to train and run AI models. The report did not disclose the size or financial terms of the planned purchase.

OpenAI has previously rented Google's chips, so the reported plan would build on an existing relationship between two companies that compete in AI products. Google is also working to expand TPU sales beyond its own services, giving cloud customers another option for AI computing. Read the full story.

3. SpaceX closes $40 billion financing deal for AI chips

SpaceX has arranged $40 billion in financing through an Apollo Global Management affiliate to acquire AI chips and server hardware, which will then be leased to the company. The leases are expected to run for slightly more than five years from the equipment's delivery date.

The deal is expected to be finalized by the end of 2026, with equipment deliveries and lease arrangements beginning in the first half of 2027. The financing allows SpaceX to secure expensive computing hardware while spreading its costs over time. Read the full story.

4. Samsung reports record profit amid the AI chip boom

Samsung Electronics reported third-quarter operating profit of 107.4 trillion won, with semiconductor profit reaching 106.5 trillion won as AI demand boosted its chip business. The company also said its high-bandwidth memory (HBM) supply for 2027 was sold out and expects demand to exceed supply further next year.

HBM is essential to many AI accelerators because it enables fast data transfers between memory and processors. Samsung plans to increase chip investment after spending 76.8 trillion won this year, while rising memory costs have also put pressure on its mobile business. Read the full story.

AI devices, infrastructure and robotics news

5. Amazon expands Alexa+ internationally

Alexa+ began rolling out in Japan on October 29 through an Early Access program, extending the assistant's availability to another major market. Japanese users can interact with it through natural conversation to handle tasks such as shopping, music discovery, calendar management, and meal planning.

Amazon plans to add integrations with services including Tabelog for restaurant reservations and Uber Eats for food delivery. Eligible customers can access Alexa+ for free during Early Access, with continued access included in Prime membership after the program ends. Read the full story.

6. Amazon announces new Echo and smart home devices

Amazon's new lineup includes the Echo Show 16, Echo Time smart clock, Alexa Photo Frame, Echo Studio with Home Camera, and a redesigned Echo Dot. The company is also adding customizable home screens and interactive controls to Echo Show devices.

Other updates include a Cards interface that saves Alexa conversations so users can revisit them across Echo Show devices, the Alexa app, and Alexa.com. The launches expand the ways users can access Alexa+ and manage connected devices throughout their homes. Read the full story.

7. Amazon introduces 3D mapping for smart home controls

Alexa Home Map creates an interactive 3D view of a home, showing compatible devices such as cameras, lights, locks, and thermostats in their physical locations. Users can select a camera to open its live feed rather than navigating through separate device menus.

The interface also provides a way to review camera activity and see smart lights alongside other connected equipment. This gives users a spatial overview of their smart home, complementing the usual list-based controls in an app. Read the full story.

8. New edge AI development board targets private, low-power computing

Coralboard, developed by Grinn and distributed by DigiKey, combines Synaptics' Astra SL2619 processor with Google's Coral neural processing unit (NPU) technology. It is designed to help engineers build and test AI applications that run directly on devices, including smart cameras, industrial equipment, and voice-enabled products.

The board supports an open-source development toolchain for deploying AI models, including hardware-accelerated support for Gemma 3 270M. Developers can use the board for prototyping before moving to a system-on-module or integrating the processor into a production device. Read the full story.

9. XREAL launches AURA XR glasses powered by Android XR

XREAL AURA uses transparent lenses to display digital content while allowing users to see the physical world. The glasses weigh less than 95 grams and offer a 70-degree field of view, while a separate computing unit houses the battery and processing hardware.

Powered by Android XR, the device supports Gemini, multiple app windows, and laptop screen mirroring through a compatible USB-C connection. Orders opened in the United States, Canada, and South Korea on October 28, with Japan scheduled to follow on November 2. Launch pricing starts at $1,099 through the end of 2026. Read the full story.

10. FIA and AWS explore AI-assisted Formula 1 car design

The FIA's partnership with Amazon Web Services will support aerodynamic simulations, technical analysis, and work on future Formula 1 regulations. One potential application is using AI models trained on existing computational fluid dynamics (CFD) results to estimate outcomes for new designs much faster than running some full simulations.

The FIA also plans to explore AI for reviewing technical regulations, analyzing vehicle data, and identifying unusual patterns related to engine performance and safety. Engineers will continue to lead the design process, with AI serving as a tool for analysis and simulation rather than generating complete car designs independently. Read the full story.