Computex 2026 Highlights Nvidias RTX Spark, Microsofts Surface Ultra, and AI-Related Legal and Security Developments
On May 31, Nvidia announced its RTX Spark superchip, an Arm‑based system‑on‑chip that marries a 20‑core Nvidia Grace CPU with a Blackwell RTX GPU and unified memory. The company and Microsoft said the chip is designed for local artificial‑intelligence, creative, and gaming workloads on Windows‑on‑Arm devices. HP confirmed that the new line of Windows PCs will be powered by RTX Spark.
Microsoft followed suit with the Surface Laptop Ultra, the most powerful Surface model to date. The 15‑inch device features a mini‑LED PixelSense Ultra touchscreen that can reach 2,000 nits of peak HDR brightness. It also incorporates the RTX Spark SoC and can support up to 128 GB of unified memory.
Dell aimed its new XPS 13 at students and young professionals. The company said the laptop is thinner and lighter than previous models, offers a larger display, more storage, faster Wi‑Fi, a touchscreen, and a backlit keyboard. Dell positioned the XPS 13 as a direct competitor to Apple’s MacBook Neo.
In the gaming arena, Asus introduced the ROG Edition 20 lineup and the ROG Xbox Ally X20, while Intel and AMD also presented new processors and related technologies.
Apple’s product roadmap also received attention. Bloomberg’s Mark Gurman reported that Apple may postpone the launch of its AI‑powered smart glasses to late 2027, a delay that follows development challenges. Gurman also cited rumors of the iPhone 18 Pro and Pro Max, which would feature a variable‑aperture main camera, aluminium frame colour options, and a slimmer Pro Max.
AI‑related legal and security news dominated the week. The state of Florida filed a lawsuit against OpenAI and CEO Sam Altman, alleging that the company released ChatGPT without adequately addressing safety risks and that it violated product‑liability laws. The lawsuit claims that OpenAI ignored warnings about potential harm.
Security incidents involving AI also surfaced. Meta’s AI chatbot was reportedly used by attackers to compromise high‑profile Instagram accounts, including the former White House account of Barack Obama and the beauty retailer Sephora. The incident illustrates the potential for AI tools to be misused for social‑engineering attacks.
Anthropic expanded access to its Claude Mythos model, adding support for around 150 new organisations in more than 15 countries, including India. The expansion is part of the company’s Project Glasswing, which aims to broaden the availability of its AI services.
The week’s events underscore the rapid pace of development in AI‑enabled hardware and the growing scrutiny of AI safety and security. While Nvidia and Microsoft push the envelope in AI‑ready PCs, Apple’s delays and the legal challenges faced by OpenAI highlight the regulatory and ethical dimensions that accompany these advances.
As the industry moves forward, stakeholders will watch how the new chips perform in real‑world workloads, how Apple’s smart‑glass plans evolve, and how the legal and security incidents influence policy and best practices for AI deployment.