Apple Releases AI Prototype Tool SQUIRE Aimed at Redesigning the UI Design Process

Apple has recently released two key studies in the field of machine learning, showcasing its latest advancements in using large language models (LLMs) to simplify UI development and enhance image security reviews.

SQUIRE: Say Goodbye to "Black Box" Design

To address the pain points of "poor controllability and difficult fine-tuning" in the AI-generated UI process, Apple introduced a system called SQUIRE. This system is powered by GPT-4o, and its core innovation lies in the Slot-Query Intermediate Representation.

Differing from tools that directly generate static designs, SQUIRE first generates a customizable component tree. Developers can modify fonts, add layers, or replace specific elements like building blocks before delivering the final code. This "white-box" process transforms the time-consuming "prompt trial-and-error cycle" into an intuitive interactive collaboration. Once the prototype is confirmed, SQUIRE can convert it into HTML and CSS code with one click. In testing with 11 front-end developers, the tool received high praise for its ease of use and controllability.

SafetyPairs: Giving Visual AI a "Protective Suit"

In terms of image generation safety, Apple proposed the SafetyPairs framework. By using counterfactual image generation technology, researchers created 1510 pairs of extremely similar images that differ in key features (for example: one image shows a building normally, while another shows the building on fire).

This dataset aims to precisely identify weaknesses in visual language models' safety ratings. Through this "spot the difference" type of training, Apple is able to develop more efficient protective models. This technology is expected to further enhance the safety level of local AI tools such as Image Playground on iPhone.

In the latest Xcode 26.3, Apple has already introduced support for agentic coding tools. The industry generally expects that the component-based generation approach represented by SQUIRE is highly likely to be officially integrated into Apple's official toolchain at the WWDC2026 event on June 8th.

World's First! China Launches 'Panzhi Yuhen Carbon Accounting Model': Accurately Portraying Global Carbon Footprints

The Shanghai Institute of Advanced Technology, Chinese Academy of Sciences, launched the world's first all-round carbon emission accounting system, 'Panzhi Yuhen Carbon Accounting Model', achieving a technological breakthrough from 'following the trend' to 'reconstructing the paradigm'. This system breaks through the barriers of traditional carbon accounting through the integration of data, algorithms, and computing power, solving bottlenecks such as high knowledge barriers, slow data updates, and low resolution, and building a solid underlying support system.

Alibaba AI Architecture Restructuring! Fei-Fei Li Appointed as CTO of Alibaba Cloud, Tongyi Lab Promoted to Large Model Business Division

Alibaba announced an organizational restructuring, with the core focus on accelerating AI development. CEO Wu Yongming announced in an internal letter the establishment of the Group Technology Committee and the upgrading of business departments, marking the beginning of a fully accelerated AI era. The most attention-grabbing news is the joining of global top scientist Fei-Fei Li as CTO of Alibaba Cloud, who will be responsible for all technical aspects and AI cloud infrastructure of Alibaba Cloud.

Li Shubin of Meituan: The First Step for Catering Businesses to Embrace AI Is Information Digitization

At the 2026 China Catering Chain Summit, Li Shubin of Meituan emphasized that the catering industry should first strengthen the local lifestyle information infrastructure when embracing AI, rather than blindly pursuing technological upgrades. He pointed out that AI does not truly understand the physical world, while consumer needs are shifting from simple searches to complex multi-intent expressions.

Google Search AI Overview Accuracy is Only 90%, Easily Affected by False Information

According to The New York Times, the accuracy of Google's AI Overview feature is about 90%. With Google's annual search volume exceeding 5 trillion searches, this means that millions of incorrect answers may be generated every hour, and nearly a million pieces of incorrect information per minute. An assessment by startup company Oumi showed that the accuracy of Google's Gemini model increased from 85% in October last year to 91% in February this year.