Search AI Products and News

Apple's Safari 26.6.1 update fixes 22 WebKit-related CVE vulnerabilities, of which 9 were discovered and assisted in resolving by OpenAI, accounting for about 41%. The key tool is OpenAI Codex Security, an AI application security agent designed for engineering and security teams.

OpenAI commits to large-scale deployment of NVIDIA AI infrastructure by 2030, building a world-class AI factory at PORTS-Pike using NVIDIA's full-stack DSX AI factory platform, spanning GPU, CPU, networking, and infrastructure software, underscoring a new surge in compute infrastructure.....

NVIDIA is discussing with SB Energy, a clean energy company supported by SoftBank, to invest up to 3 billion USD as a core part of its participation in the OpenAI data center transaction. The move aims to accelerate the construction of AI infrastructure and deeply support the AI industry chain with substantial financial power.

Apple faces sales pressure in China and is adjusting its AI strategy. It has trained a dedicated large language model for the Chinese market, with Alibaba providing technical support for core training. Previously, Apple was expected to adopt a domestic third-party large model.....

Amazon announced that OpenAI's cybersecurity models Daybreak Red and Daybreak Blue are officially available on Amazon Bedrock, now accessible to eligible users. This move deepens their long-term strategic cooperation and continues to expand Amazon's cloud AI service portfolio. Both models belong to OpenAI's Daybreak network defense program, with Red primarily providing GPT-5.6Cyber access rights, and Blue covering another set of defensive capabilities.

Last week, global AI large model calls reached 69 trillion tokens, up 21.48% WoW. China's weekly calls hit 34.25 trillion tokens, up 21.76%, maintaining a strong lead; the US had 9.17 trillion. China shows a clear advantage in global AI computing and applications.....

Wuwen Xinqiong and MiniMax signed a strategic partnership focusing on four areas: jointly optimizing model inference efficiency, building large model token services and AI infrastructure, co-innovating application paradigms, and expanding industry intelligent scenarios. They complement each other: Wuwen excels in computing power scheduling and inference infrastructure, while MiniMax has self-developed large models and massive application traffic.....

Ant Lingbo Technology is planning to seek independent financing, moving from a wholly-owned incubation within the group to the external market. The main reason is that Ant Group has been continuously increasing its investment in the AI sector, and large models and computing infrastructure have been consuming resources, leading to budget pressures for internal computing power, which has driven the embodied intelligence segment to seek capital support independently.

MiniMax released open-source multimodal generation model H3, and Moore Threads rapidly adapted it for the domestic intelligent computing card MTT S5000. H3 breaks single-task limitations by fusing text, image, audio, and video inputs to accurately understand creative intent, and can generate 2K resolution videos up to 15 seconds long.....

Alipay upgrades its AI developer incentive program, offering up to 5,660 yuan in token subsidies and lowering barriers. Individual developers integrating AI pay-as-you-go, AI web, or mobile payment can benefit from now until August 21.....

Anthropic updated Claude voice mode (July 23, 2026) with Opus and Sonnet support and cross-app automation, enabling complex tasks beyond fast responses. It defaults to inheriting text chat context for coherence.....

Google Maps beta reveals Gemini-based "personal intelligence" integrating Workspace and personal data, breaking app silos. Future Maps will combine trip planning with personal info via conversational discovery, advancing Google's cross-app smart ecosystem.....

Alibaba confirmed that its self-developed large model, Tongyi Qianwen, has integrated with Apple Intelligence and will be fully deployed across China-region iPhones, iPads, Macs, and Vision Pro. Users can invoke it directly from iOS, iPadOS, macOS, and visionOS for deep image-text understanding and high-quality generation, eliminating tedious app switching.....

JD.com and Tencent announced integration of JD’s AI agent with Tencent Yuanbao mini-programs. The agent provides full-category product info and end-to-end shopping within Yuanbao chats—from search to purchase, no redirect. JD self-developed the agent, Yuanbao’s first e-commerce vertical partner, already connected to multiple phone makers’ AI agents.....

JD's AI agent and Tencent Yuanbao have linked mini-programs, making JD the first e-commerce partner on Yuanbao. The partnership covers digital, home appliances, beauty, food, etc. Users can complete search, recommendation, order, fulfillment, and after-sales on Yuanbao, closing the loop from AI chat to e-commerce.....

LibTV introduces Agent and Skill Hub, packaging hundreds of AI video workflows into callable skills. Creators input needs, and the Agent guides direction through smart Q&A, enabling modular integration from script to editing, boosting creative freedom and efficiency.....

Zhipu founder Tang Jie announced the "Gaokao" plan, a two-year strategic investment in four core engines: long-context tasks, autonomous agents, fully self-training, and ultimate safety governance, aiming for next-gen AGI. Meanwhile, GLM-5.2 was released as open source under MIT license, supporting million-token context and leading in long-range tasks.....

Tencent is negotiating to become the largest shareholder of Singapore AI startup Manus, after Meta's $2B acquisition was ordered revoked. Existing backers including Tencent, ZhenFund, and Sequoia China plan to support the revocation at the same valuation, with Tencent subscribing for the most shares and possibly retaining a minority stake.....

Meituan has internally restricted the use of Doubao, a large model under ByteDance, requiring business teams to self-check their usage. If retention is needed, they must submit a necessity explanation and migration plan. This move shows that Meituan is accelerating its shift toward self-developed AI solutions and adjusting its infrastructure choices.

The Claude series large models from Anthropic have been launched on the Microsoft Azure AI Foundry platform. Enterprise users can now directly deploy the flagship model Claude Opus 4.8 and the cost-effective Haiku 4.5 on Azure, with key features such as prompt caching and extended thinking fully retained.