A speculative narrative describes a major tech conglomerate dissolving its flagship productivity suite, absorbing the team into the AI division and sales into cloud infrastructure, signaling the end of the standalone office software era. The article argues that foundation models and autonomous AI agents can navigate file systems and execute tasks 24/7, making legacy suites defenseless. The economic model of generative AI inverts traditional scale: each additional query escalates inference costs, turning large consumer user bases from assets into liabilities. Enterprise productivity tools, however, generate high-margin recurring revenue and hold deep organizational context that external algorithms cannot replicate, leading to a strategic bundling of AI platforms with enterprise workflows. Despite AI assistance, human review remains essential for error-prone outputs, potentially shifting rather than reducing labor; meanwhile, the traditional user interface may dissolve into invisible background infrastructure.
Apple reported its fiscal Q2 2026 revenue of $109.4 billion, up 16.4% year-over-year, with net profit surging 27% to $29.8 billion, driven by 21.7% iPhone sales growth and 28.7% Mac growth. The company's market cap briefly crossed $5 trillion on July 29, making it the second to do so after Nvidia. However, shares fell more than 8% after-hours as investors pivoted to AI-focused stocks, reflecting doubts about Apple's AI narrative. On the AI front, WWDC 2026 finally launched the revamped Siri with personal context capabilities, and China's cyberspace regulator approved 'Apple Intelligent' in July, with Alibaba's Tongyi Qianwen and Baidu models integrated for local devices. These moves come as Apple faces a high sales base from the iPhone 17 super cycle, rising memory costs, intensifying AI phone competition, and a leadership transition with John Ternus set to become CEO on September 1, succeeding Tim Cook.
A report indicates that artificial intelligence companies are purchasing old books, scanning their contents, and then destroying the physical copies. The brief content does not name specific firms or provide numbers. The practice raises concerns about data sourcing and the loss of physical cultural artifacts.
A V2EX user commented on DeepSeek's release naming, noting that V4 flash has moved from a preview to an official launch, while the Pro version still lacks an official release. The user described this scheme as confusing and unfriendly for promotion. The comment also briefly thanked the creator for the DeepSeek Agent, which they found useful.
Cerebras has secured a multi-year computing power procurement agreement with OpenAI, with a potential total value exceeding $20 billion, to be delivered by 2028; the company's current wafer-scale engine WSE-3 uses a 5nm process, covers 46,225 mm², integrates 4 trillion transistors and 900,000 AI cores, peaking at 125 PFLOPS. Tesla restarted Dojo 3, relying on TSMC's InFO_SoW wafer-level packaging to integrate 25 D1 chips into a training tile. In China, a Tsinghua University team demonstrated a 12-inch wafer-scale AI chip prototype and published three ISCA papers showing up to a 3.12x average performance improvement over GPU clusters for large-model inference mapping. The Institute of Computing Technology, CAS, developed a 16-module prototype called 'Ying Tian Lake' with 1.45x and 1.78x improvements in linear algebra and inference tasks, and proposed the Ouroboros wafer-scale SRAM compute-in-memory architecture that integrates 54 GB of SRAM on a single wafer, delivering 150,000 tokens/s for Llama 13B inference. New Ziguang Group announced the 'Zixian' 3D near-memory computing architecture achieving 30 TB/s memory bandwidth and planning wafer-level integration. Wafer-level chips are moving from technical demonstrations into industrial validation.
On July 30, ByteDance merged Feishu’s product team into Doubao (its AI assistant) and its sales and service teams into Volcano Engine, ending Feishu’s run as an independent BU; Feishu CEO now reports to the Doubao head. The same day, Tencent’s WorkBuddy V5.3.5 update enabled real-time human-AI co-writing directly inside Tencent Docs, ranking it first among Chinese desktop AI office agents with 20.97 million monthly visits. Days earlier, Alibaba started internal testing of Qianwen Office, which bundles three AI agents under DingTalk’s new CEO, backed by cloud ARR exceeding 35.8 billion yuan and triple-digit AI revenue growth. All three actions target urgent monetization: ByteDance leans on subscription tiers (68-500 yuan/month) and Feishu’s 3-billion-ARR sales channel; Tencent embeds AI as a value-added module within existing paid apps; Alibaba locks in enterprises via a cloud+AI ecosystem. The moves reflect an industry consensus that standalone collaborative office suites are being absorbed as tools for AI agents, with monetization now the overriding priority.