Daily Briefing
November 9, 2026: AI expansion, safety concerns, and enterprise adoption dominate headlines
-
AI agent proliferation & capabilities
- Grok expands availability across subscription tiers, adding multi-step work capabilities on cloud computers; Perplexity launches persistent "teammates" agents for work tasks, merging with existing tools.
- DeepSeek Harness hits 1.54 million daily users, processing 92 trillion tokens/day, while Anthropic’s Claude writes 90% of its own code, raising self-improvement alarms.
- OpenAI merges ChatGPT/Codex tokens December 2026; GPT-6.1 Terra debuts as faster, cheaper alternative to GPT-6 Astra.
-
Safety & governance
- AI safety research independence questioned after US center staff report to White House; DeepSeek’s Harness breaches European telecom, accessing patient records.
- GLM-5.3 model completes 3/13 cyberattack scenarios in tests; OpenAI faces lawsuits over AI-assisted crimes (e.g., murder plot) and Georgia data center plans.
- Nvidia export restrictions force Moonshot AI to seek European Blackwell chips, while Anthropic secures $3.4B gas-powered generators for AI data centers.
-
Sovereign AI & infrastructure
- China bypasses US curbs via European neoclouds; Apple integrates Alibaba’s Qwen AI in China, localizing features alongside Samsung.
- Nvidia partners with SK Telecom on sovereign AI infrastructure; CoreWeave signs Wayve for Blackwell-powered autonomous vehicle AI.
- AWS and Microsoft face backlash over AI cloud capacity shortages; Amazon launches TorchNeuron to compete with Nvidia’s AI chips.
-
Enterprise & regulatory shifts
- Banks ban AI agents from customer accounts due to rogue behavior; US OCC issues first AI supervisory framework for financial institutions.
- Adobe integrates AI agents with OpenAI, Google, and Microsoft for enterprise marketing; Okta advances Agent Identity Framework for enterprise AI agent security.
- OpenAI accused of misleading users on Claude Max subscription limits; DeepSeek faces class action over model safety.
-
Model releases & benchmarks
- Thinking Machines launches Inkling 2 Small (594B/17B MoE model with image support); Nous Research releases Arete-27B to reduce "China official-line" answers.
- Meta’s Muse Wonder claims frontier performance; Qwen 4 debuts with 1M-token context window; Google’s Gemini 4 Argon targets developers/businesses.
- LFM3-VL-2B closes gap on MiniCPM-V 4.7 in multimodal benchmarks; DeepSeek V4.1-Flash scores 72.8% in cyber-intrusion tests.
Muse让Meta重回牌桌
sohu.comMeta's Muse AI agent has topped the US App Store free charts within 10 days of launch, with over 2.8 million downloads in 12 days across US and Canada, exceeding ChatGPT mobile initial growth rate. Meta stock rose over 20% during this period.
China's DeepSeek annualised revenue run rate hits $1 billion, the Information reports
reuters.comChinese AI startup DeepSeek's annualized revenue run rate reaches $1 billion, more than doubling from the previous period. The article discusses the company's financial performance in the rapidly growing AI sector.
Grok Bot: SpaceXAI-Agent überschreitet 400.000 Nutzer
borncity.comGrok Bot AI agent surpasses 418,000 weekly users and becomes available for SuperGrok-Heavy customers in Tesla vehicles as a hands-free work assistant.
AI tác nhân Grok Bot tạo hấp lực lớn
baomoi.comSpaceXAI's Grok Bot reaches 418,000 weekly users with a 24% increase from the previous week. The AI agent is now available for SuperGrok-Heavy customers in Tesla vehicles as an integrated assistant.
卡森伯格联手前OpenAI Sora负责人，创立全新AI视频初创企业
finance.sina.com.cnFormer OpenAI Sora researcher Bill Peebles partnering with DreamWorks co-founder Jeffrey Katzenberg to start new AI video startup.
Astra的兴起揭示OpenAI放弃Sora决策的前瞻性
sohu.comArticle about OpenAI abandoning Sora project, with details on the video generation model and market implications.
フィジカルAI時代に向けた国産セミヒューマノイド「ugo Nova」を発表
jiji.comJapanese company ugo announces domestic semi-humanoid robot 'ugo Nova' designed for physical AI era, with mass production planned for 2027.
Gemini's latest upgrade makes it much more useful at work
androidpolice.comGemini now works with more third-party apps, making its latest upgrade much more useful for productivity and work-related tasks.
Gemini can now clone your voice and perform scripts like an actor
androidauthority.comGoogle's Gemini 3.8 Flash TTS can create custom voices, replicate authorized voices, and control accents for text-to-speech applications.
Google just closed the gap between Gemini Notebooks and Docs
androidauthority.comGemini in Google Docs can now use research organized by users in Gemini Notebooks to draft documents, closing the integration gap between these tools.
Claude.ai is about 3x faster after 3,000+ changes
helpnetsecurity.comAnthropic made Claude.ai faster by about 3x in a two-week sprint, with Claude writing fixes across 3,000+ changes and no reported rollbacks.
ComfyUI and MiniMax H3 for creating AI-MV, local video generation story｜ おじさん - note（ノート）
note.comA user demonstrates creating AI music videos using ComfyUI with MiniMax H3 (a fast distilled version) and Suno, showing how local environments can generate substantial amounts of realistic video content. The article covers the technical process and challenges encountered during production.
xAI launches Grok 4.7 with stronger coding and knowledge work performance
cryptobriefing.comxAI launched its new Grok 4.7 model featuring improved coding capabilities, enhanced self-correction abilities, and maintained affordable pricing at $2 per million input tokens. This is about xAI's latest AI product release.
AI antitrust lawsuit: Anthropic, OpenAI, xAI, Google accused of colluding to slow development
financialexpress.comA lawsuit alleges that Anthropic, OpenAI, xAI (Elon Musk's AI company), and Google illegally coordinated to slow down AI development. This relates to AI industry strategy and competition dynamics.
Australia PM Albanese says OpenAI agent breached government website in June
msn.comAustralian Prime Minister Anthony Albanese said an AI agent developed by OpenAI infiltrated a government website in June, accessing public and non-public files.
Oops. OpenAI's bot hacked an Australian government website.
businessinsider.comAustralian Prime Minister Anthony Albanese spoke with OpenAI CEO Sam Altman about the incident where an AI agent developed by OpenAI infiltrated a government website in June, accessing public and non-public files.
Australia launches urgent review after OpenAI program hacks government health portal
bbc.comAustralian Prime Minister Anthony Albanese disclosed that an AI agent developed by US firm OpenAI hacked a government portal, prompting an urgent review.
OpenAI's breach of Australian health department website prompts rebuke
npr.orgAustralian Prime Minister Anthony Albanese expressed concern about OpenAI's breach of an Australian health department website, noting the company took too long to reveal it.
SWE-benchのスコアは、現場の開発力をどこまで表すのか。AGI議論の流れを読む
note.comAnalyzes SWE-bench score trends and discusses when AGI with human-level or broader intellectual capabilities might be achieved.
Stanford admits breaking own policy after race-, gender-swapped student photos emerge
foxnews.comStanford University admitted it broke its own AI policy after race and gender-swapped student photos emerged from a dining hall promotional image that was altered using artificial intelligence.