google

Google's Suncatcher prototype space data center satellite is online and communicating with the ground after launching on a SpaceX Falcon 9, running Google DeepMind's Gemma 4 on onboard TPUs to perform inference in orbit. The test will assess how the chips handle space radiation and thermal stress, with two more satellites planned for 2027.

Asvinth ILSB，据称 Larry Page 多次在 Google 内部表示「我宁愿破产也不愿输掉这场比赛」，显示出高层对 AI 竞争的决心远超外界关注的 ROI 讨论。

Jules（Google 的异步编码智能体）发布感言，提到早期实验帮助塑造了 Google 内部多个产品，并预告后续动向。

Alphabet's Q3 2026 results showed revenue growth but a sharp stock drop after Google admitted its AI Mode search queries earn less revenue per query than traditional web search, prompting plans to increase ad load to offset the shortfall.

South Korea's KMCC fined Google KRW 42.9 billion ($32 million) and Apple KRW 20.81 billion ($15.6 million) for violating the Telecommunications Business Act by forcing in-app payment methods, imposing a 26% third-party payment commission, delaying app reviews, and applying discriminatory commission calculations for Korean developers.

A critical response to Google's 2008 blog post explaining how it decides body-of-water names in Google Earth, accusing Google of caving to political pressure over the Persian Gulf/Arabian Gulf dispute and failing to transparently justify its 'local use' naming policy.

Artificial Analysis intelligence index v5 is drawing criticism for introducing private tests on private datasets that appear to favor closed models like GPT and Google's while significantly downranking open-weight Chinese models such as Kimi 3.1 Pro and DeepSeek 4.1 Flash, coinciding with Google's Carbon 4 topping the index.

A discussion post questioning whether Google's claimed use of recursive self-improvement (RSI) signals the arrival of the technological singularity, invoking Vernor Vinge's and I.J. Good's definitions.

User expresses interest in the XREAL / Google AURA XR glasses, highlighting true vision passthrough and a 70-degree field of view, while noting the $1,099 price tag as a hurdle toward a JARVIS-like setup.

Google is expanding ads within its AI-powered Search experiences to international markets as core search ad growth slows to 12% YoY, with Alphabet reporting $127.6B in Q3 revenue and its stock falling 8% after results.

Shopify's Tobi Lutke highlights an 'Expo Mode' built with Google's new HTML-in-Canvas API, claiming to be among the first to adopt the new web capability.

Alphabet reported Q3 2026 revenue of $127.6 billion, up 25% year-over-year, with CEO Sundar Pichai saying AI investments are driving the highest search query growth rate in five years, while Google Cloud revenue surged 103% to $30.7 billion.

Google发布前沿模型Gemini 4 Carbon，在编码、智能体任务和网络安全方面实现显著提升，在Artificial Analysis Intelligence Index、SEC-bench Pro和ExploitGym等多个基准上领先Claude Opus 5.5和GPT-6 Astra，定价与Argon一致。

Google 宣布推出新一代前沿 AI 模型 Gemini 4 Carbon，据称在编码、智能体任务、科学分析及网络安全漏洞检测方面有显著提升，并在多项基准测试中超越 GPT-6 Astra 和 Claude Opus 5.5，但采用分阶段发布，用户暂时无法使用。

Google and Xreal's Aura AR glasses launch at a reduced $1,099 price, undercutting Meta's VR glasses with a lightweight 95-gram design, Android XR, Gemini integration, and 150+ XR experiences.

Google has reclaimed the top spot on the Artificial Analysis AI benchmark leaderboard, prompting commentary on the impressive result and speculation about its training cost.

Google's new Carbon model debuts at the #1 spot on the Artificial Analysis leaderboard with a score of 59, marking a notable top ranking for the company.

Google is updating Chrome's autofill with AI that can pull details like flight confirmations, tracking numbers, and addresses from users' Gmail to fill out online forms, rolling out to U.S. desktop, iOS, and Android users as part of its Personal Intelligence initiative.

An analysis of why Google, despite its vast resources and AI expertise, has never attempted to acquire OpenAI and the strategic reasons behind it.

CONCUR introduces a new advantage estimator for multi-task, multi-reward RL training of LLMs that projects gated auxiliary rewards onto scores concordant with primary rewards, outperforming baselines across math, code, instruction following, and writing tasks on three models.