Feed archive

Browse the last year of updates by day. 74 in this window; filter by type or load another week.

Digests

2026-09-194 items

09-19

Qwen3.8-LiveTranslate: real-time simultaneous interpretation across 60 languages

Read original →

Model releaseAlibabaQwen

09-19

[Editorial] Chinese X is arguing one thing: after chat, agent capability is outrunning trust

Read original →

PolicyAGI Pulseagipulse editorial

09-19

[Editorial] Claude Code reads AGENTS.md: the story is agent-instruction convergence, not one fewer file

Read original →

ProductAnthropicagipulse editorial

09-19

[Editorial] What the Jev frenzy is really about: agents lack a cheap decision layer

Read original →

Model releaseTypeSafe AIagipulse editorial

2026-09-186 items

09-18

Anthropic partners with Accenture on embedded independent AI evaluation

Read original →

PolicyAnthropicAnthropic

09-18

Google AI weekly recap: Gemini 3.8 Live, Dreambeans, CC, Google Pics and AlphaGenome Atlas

Read original →

ProductGoogleGoogle AI

09-18

Meta SAM 3.1 lands on Meta Model API for detection and segmentation

Read original →

ProductMetaMeta for Developers

09-18

MiniMax open-sources Code CLI under MIT with FrontierHarness SOTA claim

Read original →

ProductMiniMaxMiniMax

09-18

MiniMax joins Singtel AI Pass for Singapore SkillsFuture learners

Read original →

ProductMiniMaxMiniMax

09-18

Qwen launches Qwen3.8-Omni-Flash, an agentic omni-modal model

Read original →

Model releaseAlibabaQwen

2026-09-175 items

09-17

Muse personal agent launches on Mac

Read original →

ProductMetaMuse

09-17

Anthropic publishes three metrics to track the pace of AI development

Read original →

NewsAnthropicAnthropic

09-17

OpenAI launches Astra for Law on GPT-6 Astra

Read original →

ProductOpenAIOpenAI

09-17

Anthropic opens Life Sciences Verification Program, unlocks Mythos for verified users

Read original →

NewsAnthropicAnthropic

09-17

Claude Projects run parallel cloud threads from one Claude Code chat

Read original →

ProductAnthropicClaude

2026-09-1613 items

09-16

Ethan Mollick：媒体聚焦 AI 生存风险，但企业与个人采用正快速日常化

Read original →

StatementEthan Mollick

09-16

Meta：Muse Code 原生支持 Windows（无需 WSL），含 x64/ARM64 与零管理员安装

Read original →

ProductMetaMeta for Developers

09-16

OpenAI：公布模型错位（misalignment）披露框架，并同步发布六例报告

Read original →

PolicyOpenAIOpenAI

09-16

Anthropic：Claude Cowork 与聊天合并为统一 Claude，并上线 Docs/Slides/Design

Read original →

ProductAnthropicAnthropic / Claude

09-16

hardmaru：Sakana AI 一年内从零组建产品团队并密集发货

Read original →

StatementSakana AIhardmaru / Sakana AI

09-16

Ethan Mollick：AI 正让组织内岗位边界模糊，需尽快重建分工模型

Read original →

StatementEthan Mollick

09-16

DeepMind launches Institute for AGI safety, governance, and societal impact

Read original →

StatementDeepMindDemis Hassabis / DeepMind Institute

09-16

Hassabis: arts and humanities will shape the AGI-era future

Read original →

StatementDeepMindDemis Hassabis

09-16

OpenAI：ChatGPT Work / Codex 管理后台上线用量—成本

Read original →

ProductOpenAIOpenAI

09-16

LeCun: better AI is safer AI; pacing is counterproductive

Read original →

StatementMetaYann LeCun

09-16

LeCun: formal proofs are not superhuman math

Read original →

StatementMetaYann LeCun

09-16

【待核实】据报道 OpenAI 通过承包商审阅 ChatGPT 对话（Project Lily）

Read original →

NewsOpenAIX news story / 媒体转述

09-16

【待核实】OpenAI 多智能体声称攻克纳维

Read original →

NewsOpenAIX news story / 待核实传播

2026-09-1519 items

09-15

Zuckerberg: alignment is becoming a key model capability

Read original →

StatementMetaMark Zuckerberg

09-15

TypeSafe AI launches System One model Jev with RLCD for structured decisions

Read original →

NewsTypeSafe AITypeSafe AI / Diogo Almeida

09-15

Brockman: ChatGPT Work can act on company data and dashboards

Read original →

ProductOpenAIGreg Brockman

09-15

Pichai: Google AI for science, AlphaGenome Atlas and more

Read original →

StatementGoogleSundar Pichai / Google

09-15

Google DeepMind launches Gemini 3.8 Live and Live Extended Thinking

Read original →

NewsGoogleGoogle DeepMind

09-15

Odyssey unveils Odyssey-3 foundation world model

Read original →

Model releaseOdysseyOdyssey

09-15

Salesforce in Claude is now in beta

Read original →

NewsAnthropicClaude

09-15

Sam Altman：本周还有大批发货，DevDay 还会再加一波

Read original →

StatementOpenAISam Altman

09-15

什么样的数据库底座，能加速 Agent 从 Demo 走向生产？

Read original →

News

09-15

Hugging Face says it was a disclosed agent cyberattack victim

Read original →

StatementHugging FaceClement Delangue

09-15

Anthropic：Claude for Small Business 新增工作流与 27 项集成，覆盖获客到结账

Read original →

ProductAnthropicAnthropic / Claude

09-15

Meta 打造“组织第二大脑”智能体的设计思路

Read original →

NewsMeta

09-15

从 Trace 到规模化实时评估：面向生产流量的 Agent 可观测实践｜QCon上海

Read original →

News

09-15

Musk admits Grok 4.7 has not surpassed rivals, teases AGI with Grok 5

Read original →

StatementxAI디지털투데이

09-15

前OpenAI后训练VP回应陶哲轩：说AI毁了数学，可能还是太小瞧AI了

Read original →

NewsOpenAI

09-15

7名博士生仅用3个月从零训练7B大模型：代码+数据+训练日志全公开

Read original →

News

09-15

Ex-FTC boss Khan: break out the handcuffs for AI CEOs, citing 1934 precedent

Read original →

News

09-15

Google Faces Red Flags in AI Development as Doubts Grow Over Delayed High-Performance Model Release

Read original →

ProductGoogleasiae.co.kr

09-15

AI 负责创造，人来干脏活，这事儿能否停一下？

Read original →

News

2026-09-1421 items

09-14

大模型重压之下

Read original →

News北京商报

09-14

发长文预警 AI 风险后，Dario 首次专访回应：AI 不能停，但必须慢下来

Read original →

News

09-14

Arm 推出 AI Portal：当 AI 应用从“模型可用”走向“平台可用”

Read original →

Product

09-14

Pion, an agent designed to run any company autonomously

Read original →

News

09-14

大模型“压力测试周”：DeepSeek调价、Kimi辟谣、智谱融资

Read original →

NewsDeepSeek新浪新闻_手机新浪网

09-14

不下班的经营者：把 LLM-as-Judge 做成会自我校准的评估闭环｜QCon上海

Read original →

News

09-14

OpenAI may have violated California’s AI safety law with latest model releases, AI watchdog says

Read original →

PolicyOpenAIFortune

09-14

Dropping eBPF CPU Cost by About 90% with Memoization (Not AI Gen)

Read original →

News

09-14

openJiuwen 首发双维度 RSI 框架，AI 自修改，落地办公智能体，算力亲和助力又快又省

Read original →

News

09-14

Shanghai AI Lab releases Atria Dawn Preview: 744B MoE agentic model (MIT)

Read original →

NewsShanghai AI LabAtria ASI / InternLM

09-14

GitLab 警告称，AI 代理的沙箱安全性取决于其网络访问的安全性

Read original →

News

09-14

国内首个团队智能体“豆包工作伙伴”发布_财经_24H快讯

Read original →

ProductByteDancebihainews.com

09-14

OpenAI bots knew about the RubyGems caching vulnerability

Read original →

NewsOpenAI

09-14

Why DeepSeek-V4.1-Flash Is Such an Exciting Open Model Release

Read original →

Model releaseDeepSeekKDnuggets

09-14

首届蚂蚁灵波具身大模型挑战赛正式启动

Read original →

News

09-14

探索RSI，生数新世界模型让机器人开始自我进化

Read original →

News

09-14

分子之心QuantaMind登Science Advances，让AI给分子世界”拍电影”

Read original →

News

09-14

端侧AI从「能跑」到「会进化」，元空智能跑进惠普预装

Read original →

News

09-14

三个月价格腰斩，大模型的“聪明”正在贬值？

Read original →

News21财经

09-14

豆包手机助手发布消费者版本，行业首款AI智能体手机9月16日开售！

Read original →

NewsByteDancet.cj.sina.cn

09-14

“豆包手机助手”消费者版本正式发布 - OSCHINA - 开源 × AI · 开发者生态社区

Read original →

ProductByteDanceOSCHINA

2026-09-136 items

09-13

蚂蚁发布大模型内生式安全护栏SingProbe，让AI边生成边识别风险

Read original →

News环球网

09-13

“Kimi时刻”到来！月之暗面赶赴大模型上市潮

Read original →

NewsMoonshot新浪网

09-13

OpenAI CEO Sam Altman Warns of ‘A Complete Change in the Landscape of Cyberattacks’ as AI Models Around the World Hit ‘Cyber Critical’

Read original →

StatementOpenAIBarchart.com

09-13

DeepSeek下一波大模型被指即将问世：3万亿参数性能超GPT-6 Astra

Read original →

NewsDeepSeek驱动之家

09-13

Figma 如何利用 AI 代理提升安全性

Read original →

News

09-13

Read, Don't Write: 重塑大模型评价体系，构建全自动、可进化的“探测式”评测管线｜QCon上海

Read original →

News