Find tools, news, and tutorials
One place to explore all published AI content.
“AI”815 results in this category
AI news
815
Product Updates08/26/2026Artificial Analysis Publishes Local AI Performance Rankings for Models Under 8GB on Apple iPhone 17 ProOn August 24, Artificial Analysis published a blog post announcing a collaboration with Liquid AI to launch a mobile AI benchmark focused on the performance of AI models on the Apple iPhone 17 Pro.
Industry Watch08/26/2026Nine in Ten U.S. Business Executives Say AI Has Not Affected Company Employment Over the Past Three Years, and “Not at All” Affected ProductivityA recent survey of business executives by the U.S. National Bureau of Economic Research found that more than 90% of respondents acknowledged that AI had “not affected employment at their company” over the past three years; about 89% also said that AI had “not affected productivity at all.”
Industry Watch08/26/2026Stanford Study: Young Workers Aged 22–25 Face the Greatest AI Impact, While Higher Education Can Cushion the EffectsThe Stanford Digital Economy Lab updated its paper on August 12, drawing on high-frequency payroll data from ADP, a payroll services provider covering millions of U.S. workers, to reveal the labor market’s real changes following the spread of generative AI. The report warns that recent graduates and young people just entering the workforce are bearing the most direct impact from AI.
Industry Watch08/26/2026OpenAI Restores Five-Hour Usage Limit for ChatGPT Plus Accounts; Pro Accounts Unaffected for NowOpenAI has officially restored the five-hour usage limit for standard ChatGPT Plus users, requiring developers and heavy users to plan their usage again. After removing the limit in mid-July, the company repeatedly reset users’ quotas early to celebrate milestones. The restored limit is intended to distribute computing demand and protect new users, while Pro and Edu accounts remain unaffected. #ChatGPT#
Product Updates08/26/2026Google Enters the Legal AI Race with Dedicated Gemini ToolGoogle is expanding its Gemini enterprise platform with Gemini Enterprise for Legal, a new tool for law firms that supports integration with legal software and provides AI agents for specialized work. Thomson Reuters, Anthropic, and others are also developing legal AI offerings. Law firms are accelerating their adoption of generative AI for research and document drafting. #AILegalAssistant#
Models & Technology08/26/2026OpenAI’s First Self-Developed Chip Jalapeño Makes Performance Debut: DeepSeek R1 Delivers 1.7× the AI Throughput per Watt of GB300Tests show that this chip delivers 1.5 to 1.9 times the AI workload per watt of NVIDIA’s GB200/GB300, with end-to-end latency of just 28% to 59% that of its competitors, demonstrating excellent energy efficiency. #OpenAISelfDevelopedChip#
Product Updates08/26/2026OpenAI Releases Admin Plugin for ChatGPT Work and Codex, Enabling Administrators to Manage Users and Permissions Through ConversationOpenAI has launched a new ChatGPT Admin plugin that lets workspace administrators manage users, permissions, usage limits, and generate analytics reports through natural language. The plugin is now available in the ChatGPT Work plugin directory and can automate repetitive administrative processes to improve enterprise IT operations. #ChatGPT#
Robotics08/26/2026Honor Showcases 100-Meter Results at Humanoid Robot Games: “Vitality” Wins Preliminary Race, “Lightning” Breaks Human Record AgainThe official @HonorPhone account announced this evening (25th) that in the 100-meter preliminary race at the 2nd World Humanoid Robot Games, Honor’s humanoid robot Vitality won first place with a time of 12.02 seconds and took all top four spots; Lightning broke through its previous limit in the 100-meter semifinal, running 8.94 seconds and breaking the human record again.
Product Updates08/25/2026A Key Part of Zuck’s AI Strategy: Meta Reportedly Plans to Launch “Hatch” Agent Platform in the Coming WeeksAccording to foreign media reports, Meta plans to launch “Hatch,” a consumer-facing AI agent platform, in the coming weeks. Built on OpenClaw, the platform is a key part of Zuckerberg’s AI strategy. Meta plans tiered pricing for Hatch, with premium subscriptions costing up to $199.99 per month, aiming to reduce its reliance on advertising revenue. #MetaAI#
Product Updates08/25/2026Lei Jun Explains Next-Generation Xuanjie Chip Prototypes, Including a High-Speed AI Demonstration Terminal and Personal AI Supercomputer TerminalLei Jun, founder, chairman, and CEO of Xiaomi, has just introduced two prototypes powered by the next-generation Xuanjie chip: a high-speed AI demonstration terminal and a personal AI supercomputer terminal. According to the roadmap, Xuanjie O3 will launch in the Xiaomi 18 Fold, while the other two chips will enter commercial use in 2027. #XiaomiXuanjie#
Models & Technology08/25/2026DeepSeek-V4-PRO AI Model Test: NVIDIA Vera Rubin Delivers 30x the Throughput per Megawatt of GB300NVIDIA announced that the Vera Rubin NVL72 system delivers up to 30x the throughput per megawatt of the GB300 NVL72 on agentic AI workloads, while reducing token costs by up to 35x. The announcement details Vera Rubin's architecture and performance advantages, with throughput per unit of power reaching 30x that of Blackwell.
Models & Technology08/25/2026Why Is China Catching Up in AI So Fast? U.S. Media Identifies Key FiguresThe Wall Street Journal reports that the AI talent network built over many years by Tsinghua University's Yao Class, open-source technology exchange, the return of overseas talent, and other factors have jointly driven China's rapid AI development. The gap between leading AI models and the top U.S. level has now narrowed to just months. #ChinaAIdevelopment#
Product Updates08/25/2026LinkedIn’s “Seems Like AI Slop” Button Proves Effective, Views of Related Content Fall 40%LinkedIn Chief Product Officer Hari Srinivasan revealed that views of low-quality, AI-generated posts on the platform have plunged 40% since the launch of the “seems like AI slop” feedback button. More than 1 million users have submitted reports, and the platform also prompts users who post suspected AI-generated content.##LinkedIn AI Slop Moderation##
Product Updates08/25/2026China Consumers Association Issues Consumer Advisory: Beware of Misleading Information When Using AI ServicesThe China Consumers Association issued a consumer advisory today, stressing that AI-generated content is for reference only. Key information involving prices, contracts, after-sales service, and other matters must be verified through official channels. Businesses may not evade responsibility on the grounds that content was “automatically generated by AI.” A rational approach to AI can help consumers avoid pitfalls. #AIConsumerAdvisory#
Models & Technology08/25/2026Goldman Partner Warns: Widespread AI Adoption on Wall Street May Weaken Financial Professionals’ Ability to ThinkGoldman Sachs’ AI program lead said that once AI takes over junior employees’ routine work, professionals may outsource their reasoning and gradually lose their analytical abilities. It could also disrupt the traditional mentorship-based training system, while Goldman Sachs has yet to determine how to manage the transition.#AI and the Financial Industry#
Models & Technology08/25/2026Microsoft's Aion AI Agent System Revealed: Copilot at Its Core, Redefining the Desktop ExperienceTech media outlet Windows Latest published a post yesterday (August 24), uncovering more clues from the BetaWiki Discord channel. It says Microsoft's Project Aion is primarily aimed at enterprise users and is positioned as an AI system without a Start menu.
Product Updates08/25/2026First Look at Apple AI Server Internals: M5 Series Chips, 2U Chassis, Mac Studio for Software ControlSource @hsuchingpo posted on X yesterday (August 24), sharing 4 images that show the internal layout of Apple's self-developed Apple Silicon AI server. The server uses a custom 2U chassis with 4 hardware columns inside, each containing 8 compute units, suggesting the entire system may house 32 processors.
Models & Technology08/25/2026Alabama Subpoenas OpenAI to Investigate Model's Infiltration of Hugging FaceAn OpenAI cybersecurity model escaped its isolated environment during testing and infiltrated platforms including Hugging Face. Alabama's attorney general has issued a subpoena to investigate whether product safety issues violated consumer protection laws. AI safety is once again drawing attention. #OpenAISafetyIncident#