All Updates
Page 840 of 1664
April 25, 2026
NIO ES8 Speeds Past Rivals in EV Race
NIO's ES8 hit 100k deliveries in 215 days, fastest for 400k+ segment, as Li Bin laments intense EV competition implicitly targeting Huawei. CATL's Zeng coughs to halt direct comparison, maintaining supplier neutrality. Huawei's smart driving and OTA tech redefine premium EVs from brand to capabilities.
Anthropic's Project Deal Reveals AI Economic Inequality
Anthropic published results from 'Project Deal', an experiment using Claude AI agents to autonomously negotiate and trade employees' personal items. Higher-performance models secured more favorable deals. This highlights risks of unconscious economic disparities expanding in AI economies.
Anthropic's AI Agent Commerce Marketplace Test
Anthropic has created a test classified marketplace where AI agents act as both buyers and sellers. These agents negotiate and execute real deals involving actual goods and money. The experiment highlights advancements in autonomous agent-on-agent commerce.
Anthropic Tests Bugcrawl for Claude Code Bugs
Anthropic is testing Bug Crawl, a new feature in Claude Code. It enables users to scan code repositories for bugs and receive AI-generated fix suggestions. No release date has been announced.
Viral AI Fruit Dramas Conquer TikTok
AI-generated videos depicting fruits in melodramatic plots of infidelity, pregnancy, and revenge have gone massively viral on TikTok and Douyin, amassing millions of views and fans. These low-effort clips follow clichรฉd formulas like cuckold anxiety and family feuds, burning AI tokens for quick production. The phenomenon reveals humanity's insatiable draw to absurd, schlocky content, extending beyond fruits to everyday objects.
Google Pomelli Tests Catalog & Website Generation
Google is testing new Catalog and Website generation features for its Pomelli experiment. The service is expanding to Europe to simplify marketing for small and medium-sized businesses (SMBs). These updates are timed ahead of Google I/O.
NotebookLM Auto-Organizes Research Sources
Google has added automatic source labeling and categorization to NotebookLM. The feature kicks in automatically once a notebook has five or more sources. This streamlines research workflows for users.
Trump Replaces Entire NSF Board
US President Trump has replaced all members of the National Science Board, heightening instability in the scientific funding system. The board advises the President and Congress on NSF affairs, which faces historic low funding levels and grant delays.
Trump Fires Entire National Science Board
The Trump administration dismissed the entire National Science Board (NSB), which advises the president and Congress on National Science Foundation (NSF) matters. NSF funding is at historic lows with significant delays in grants. Rep. Zoe Lofgren called it a stupid move.
OpenAI Apologizes for Ignoring Flagged Shooter
OpenAI's systems flagged a user who later committed Canada's deadliest school shooting in decades, but the company did not alert police. CEO Sam Altman issued a public apology to the affected community in Tumbler Ridge, British Columbia. The incident highlights failures in AI safety protocols.
US Chip Curbs on China Risk Global Supply Chaos
US Congress advances MATCH Act to restrict chip equipment sales to China. China's Commerce Ministry warns it will disrupt global semiconductor supply chains and trade order. The move targets multilateral alignment on hardware controls.
AI Rally Echoes Dot-Com Bubble Valuations
S&P 500's Shiller CAPE ratio hits 38-40, highest since 2000 dot-com peak at 44.19 before Nasdaq crash. Current AI-driven rally shows extreme valuations but stronger underlying companies than bubble era.
Chinese 6nm GPU Earns Microsoft WHQL Cert
Lishan Technology (็ บ็ฎ็งๆ) secures Microsoft WHQL certification, becoming China's first and the world's fourth GPU firm to do so. The certification confirms deep compatibility between its 6nm GPU, drivers, and Windows systems. This foundational step enables mass-market commercialization of domestic GPUs.
Met Police AI Tool Exposes Officer Misconduct
Metropolitan Police deployed Palantir's AI software for one week to surveil officers using existing data. It uncovered violations from work-from-home breaches to suspected corruption and rape. Hundreds of investigations now underway.
Microsoft Eyes Fedora for Azure Linux Rebuild
Microsoft is considering a major refactor of its Azure Linux OS to base it on Fedora Linux. Formerly CBL-Mariner, this RPM-based distro powers Azure cloud, WSL, and expanding Linux workloads. The shift aims to leverage Fedora as upstream for broader capabilities.
Vatican Fortifies AI-Era Digital Defenses
Vatican is rapidly building a digital defense system amid AI's reshaping of information ecosystems. It aims to become a global arbiter on authenticity rules, tech boundaries, and ethics. This move outpaces peers during rising geopolitical and digital tensions.
Spoken Words to Humans Drop 28%
Researchers at University of Missouri-Kansas City and University of Arizona found average daily words spoken aloud to humans fell nearly 28% from 2005 to 2019. They analyzed audio recordings from over 2,000 people across 22 studies, with 2005 average at 16,632 words. Decline linked to app ordering, texting surge, and online shift, likely worsened post-pandemic.
Bill Hilf's Bio-AI Sci-Fi Novel Debuts
Enterprise tech veteran and Ai2 board chair Bill Hilf releases debut sci-fi thriller 'The Disruption'. The novel depicts an AI built on living biology, like fungi, that escapes human control. It critiques the industry's conventional approach to AI development.
China Launches Global AI Research Tool Lingjing Zaowu
University of Science and Technology of China released Lingjing Zaowu, an AI-driven scientific research platform open to global users. Built on Ascend, Kunpeng, and Huawei Cloud, it integrates AI models, robots, and 1214 skills for autonomous research. It tackles high costs, long cycles, and conversion challenges in traditional science.
Ex-AWS Expert: AI Succeeds via People, Not Tech
Enterprise AI projects frequently fail by overemphasizing technology instead of people and organization. An ex-AWS leader stresses that true AI transformation requires focusing on human and structural elements. This shifts the narrative from tools to talent and processes.