| Time | Entity | Sector | Event / capture |
|---|---|---|---|
| 4:28 PM ET | ANAnthropic | 4:28 PM ETANAnthropic @MParekh on X ‘Nothing to See Here’ in AI, Says DC. Anthropic, OpenAI & Google. ARD #155 ...Commerce clears Anthropic. DOJ backs OpenAI on fair use. Google keeps its ad stack. Ternus up next at Apple Sep 9. AI Ramblings Daily podcast #155 on AI-RTZ. $GOOG Detailed writeup and source links: https://t.co/FRsrX7vCMd | |
| 4:51 AM ET | $GOOGL | 4:51 AM ET @tchsignal on X Gemini 3.8 Flash Puts Google's Rapid AI Release Cycle to the Test Google has released Gemini 3.8 Flash, its third Flash release in six weeks and roughly three weeks after Gemini 3.7 Flash. The new model is generally available with a 1 million-token context window, up to 64,000 output tokens, and low, medium, or high thinking levels. For developers, the important detail is not just Google's rapid release pace. Gemini 3.8 Flash keeps the same introductory per-token pricing as 3.7 Flash: $0.75 per million input tokens and $3.75 per million output tokens through December 31, 2026. But equal token… | |
| 7:54 PM ET | OPOpenAI | — | 7:54 PM ETOPOpenAI @stretchcloud on X The open-source alternative to Claude Managed Agents just launched. Omnara handles the production scaffolding your agent needs: durable execution, sandboxes, permissions, secrets, streaming, approvals.The insight in the positioning is right: most people building agents want to own the agent logic. They don't want the hosting burden. Claude Managed Agents (and equivalents from OpenAI, Google) bundle those two concerns together. Omnara unbundles them.The category is moving fast. Alongside Omnara there's OpenMA, open-managed-agents, and TrueForge from TrueFoundry doing similar things. The… |
| 6:34 PM ET | $GOOGL | — | 6:34 PM ET @stretchcloud on X My read on Muse Spark 1.3: "almost too cheap to meter" is doing real work as a positioning statement.The benchmark picture: 75.4 on DeepSWE v1.1, 88.8 on Terminal-Bench 2.1, 98.5 on MRCR 256K-512K long-context. That last number matters more than the coding scores right now. Most agentic coding tasks need to hold a full repo in working context. Muse Spark leads the long-context category against Opus 5 and GPT-5.6 Sol.The 20% tool call reduction and 25% token usage reduction vs Muse Spark 1.2 compounds fast across multi-step agent plans. Google Agentspace, AWS Bedrock Agents, LangChain, and the… |
| 5:13 PM ET | ANAnthropic | 5:13 PM ETANAnthropic @MParekh on X ‘New and Improved’ in AI Models, Anthropic, OpenAI & Google. ARD #154 ...Fable 5.1’s fine print. Astra tips out on cyber. Gemini narrows the gap. World models level up. Full writeup and source links: https://t.co/UZSXoRFmE2 https://t.co/UOEqsFebFq | |
| 4:20 PM ET | $GOOGL | 4:20 PM ET @stretchcloud on X The model tier split just happened. Gemini 3.8 Flash and Gemini 3.8 Flash Cyber launched today. Same day, completely different access regimes. The public Flash model is on the Gemini API at $0.75 per million input tokens. It scores 73.7% on DeepSWE 1.1, leads Harvey's Legal Agent Benchmark, and is the third updated Flash in 6 weeks. Gemini 3.8 Flash Cyber is not on the public API at all. Access is through Google's Fairwind Program, restricted to government agencies, critical infrastructure operators, and software maintainers. The capability gap is not marginal. Google's Chrome Security team… | |
| 11:11 PM ET | — | 11:11 PM ET @Avin_lucia on X @Arifx0001 @BitRobotNetwork Great synergy leveraging high-end and affordable tools for teleop to scale AI efficiently | |
| 10:38 PM ET | $GOOGL | 10:38 PM ET @OdinsDeposition on X @dotkrueger They essentially pinpointed their long running targets today with Nvidia and Spacex CEOs both aligned on listing 10x economic gains from robotics. Prior to that 20-30% global economic gain from widespread digital AI. | |
| 1:29 PM ET | $GOOGL | 1:29 PM ET Google DeepMind on X: "We’re bringing agentic video understanding to our latest Gemini models.
They can now analyze videos with better accuracy while using up to 88% fewer tokens. 🧵" / X Post Log in Sign up Post Google DeepMind on X: "We’re bringing agentic video understanding to our latest Gemini models. They can now analyze videos with better accuracy while using up to 88% fewer tokens. 🧵" - Google DeepMind @GoogleDeepMind We’re bringing agentic video understanding to our latest Gemini models. They can now analyze videos with better accuracy while using up to 88% fewer tokens. 🧵 View media 1:29 PM · Sep 1, 2026127.2KViews 93 132 1.3K 264 - Google DeepMind @GoogleDeepMind 21h Instead of scanning an entire file, Gemini reasons across the video’s transcript, audio, and… | |
| 10:31 PM ET | $GOOGL | 10:31 PM ET smartfitguide on X: "Google DeepMind's RT-2 model treats robot actions as text tokens. By co-fine-tuning pre-trained Vision-Language Models with robotic data, it achieved nearly double the performance on novel tasks compared to its predecessor, RT-1. #AI" / X Post Log in Sign up Post smartfitguide on X: "Google DeepMind's RT-2 model treats robot actions as text tokens. By co-fine-tuning pre-trained Vision-Language Models with robotic data, it achieved nearly double the performance on novel tasks compared to its predecessor, RT-1. \#AI" - smartfitguide @bareani21645 Google DeepMind's RT-2 model treats robot actions as text tokens. By co-fine-tuning pre-trained Vision-Language Models with robotic data, it achieved nearly double the performance on novel tasks compared to its predecessor, RT-1. #AI View media 10:31 PM · Aug 31, 20268Views Reply 1 | |
| 9:50 PM ET | $GOOGL | 9:50 PM ET 大仏ヨヨ on X: "Gemini Robotics ER 2を試してみたけど結構いいね。リアルタイムでOpenDuckMiniを追跡させてみたけどレスポンスは想定していたより悪くはなかった。どういうロボットかまで判別できているのが良かった。これ以外の機能もあるのでもう少し深堀しようと思う。
#OpenDuckMini,#microduckと同じ作者" / X Post Log in Sign up Post 大仏ヨヨ on X: "Gemini Robotics ER 2を試してみたけど結構いいね。リアルタイムでOpenDuckMiniを追跡させてみたけどレスポンスは想定していたより悪くはなかった。どういうロボットかまで判別できているのが良かった。これ以外の機能もあるのでもう少し深堀しようと思う。 \#OpenDuckMini,\#microduckと同じ作者" - 大仏ヨヨ @U6kcXA6yRN47345 Gemini Robotics ER 2を試してみたけど結構いいね。リアルタイムでOpenDuckMiniを追跡させてみたけどレスポンスは想定していたより悪くはなかった。どういうロボットかまで判別できているのが良かった。これ以外の機能もあるのでもう少し深堀しようと思う。 #OpenDuckMini, #microduckと同じ作者 00:00 View media 9:50 PM · Aug 31, 2026721Views Reply 1 12 4 | |
| 8:00 PM ET | $GOOGL | 8:00 PM ET Introducing Agentic Video in Gemini Introducing agentic video understanding with Gemini Sep 01, 2026 \| 7 min read - x.com - Facebook - LinkedIn - Mail - Copy link Our new agentic feature for video analysis cuts token consumption by up to 88%, reduces costs by up to 66%, and boosts quality by up to 7%. Rohan Doshi Senior Product Manager, Google DeepMind Mario Lučić Research Director, Google DeepMind Share - x.com - Facebook - LinkedIn - Mail - Copy link Your browser does not support the audio element. Listen to article \[\[duration\]\] minutes This content is generated by Google AI. Generative AI is experimental… | |
| 3:11 AM ET | $TSLA | 3:11 AM ET Humanoid Robots News | AI Breakthroughs, Robotics Trends & Synthetic Updates Humanoid Press - Humanoid News - Humanoid Daily - News Archives - Humanoid TV - History Board - AI News - 2026 - Top 10 Artificial Intelligence - About AI - About AGI - Workshop - Operation Instructions - Humanoid & AI Safety - Quantum Robotics - Database - Supplier Directory - CubeMars - Shopping Guide - HPS Leadership - HPS Field Test Audit Checklist - Top 10 Humanoid Company 2026 - Top 10 Humanoid What Each Does Best 2026 - Partnerships - Interview - Opinion - Magazine - About - Copyright Notice - Cookie Policy - Disclaimer - AI & Automation - Licensing & Attribution Policy - Terms of Use… | |
| 8:00 PM ET | $GOOGL | 8:00 PM ET Gemini Robotics ER 1.6 shutdown date Aug 31, 2026; replacement ER 2 Google Developers Gemini API deprecations page, Robotics models table (page last updated 2026-08-27 UTC; shutdown date in the table is in-window): gemini-robotics-er-1.6-preview released April 14, 2026, shutdown date August 31, 2026, recommended replacement gemini-robotics-er-2-preview. Older gemini-robotics-er-1.5-preview listed shutdown April 30, 2026. Companion overview (ai.google.dev/gemini-api/docs/robotics-overview, scraped Sep 1) says upgrade by replacing model='gemini-robotics-er-1.6-preview' with gemini-robotics-er-2-preview or gemini-robotics-er-2-streaming-preview; ER 1.6 'will be shut down at the end of August.' Pricing page (same scrape): ER 2 Preview paid $2.00 input / $10.00 output per 1M tokens (text/image/video/audio); batch $1.00 / $5.00; ER 2 Streaming $2.00 / $10.00; ER 1.6 Preview $1.00 input ($2.00 audio) / $5.00 output. This is an API lifecycle event, not a factory-hours print. Gemini Robotics 2 VLA launch itself was July 30, 2026 (Hassabis) — outside. | |
| 8:00 PM ET | 8:00 PM ET Apptronik Forge tape Aug 31: $36.97 (-7.94%), $5B mark; $964.7M total funding Forge Global Apptronik private-share page (Forge Price updated Aug 31, 2026): Forge Price $36.97, -$3.19 (-7.94%); Forge Price valuation $5B. Total funding listed $964.7M. Funding table: Series A-X1 02/11/2026 $441.21M at $36.97/share, $5B post-money; Series A-X2 02/11/2026 $78.79M at $36.97, $5B post-money; investors named AT&T Ventures, B Capital, Google, John Deere, Mercedes-Benz Group, Peak6, Qatar Investment Authority. Earlier Series A-1/A-2/A-3 03/18/2025 totaling about $415M at $1.78B post-money. Active-market widget timestamped 8/27/2026 2:33:46 PM EDT (3 bids, total bids $215.95K). News module still points at Robot Park coverage dated 06/30/2026 Forbes and 07/06/2026 Robotics & Automation News — those factory items are outside this window; the in-window fact is the Aug 31 private mark and listed capitalization, not a new Robot Park opening. | ||
| 8:00 PM ET | 8:00 PM ET Xpeng robotics unit: >$900M at >$6.3B; Iron mass production targeted end-2026 Vytautas Valinskas / Technology.org (Aug 31, 2026), largely recapping TechCrunch Aug 28: Xpeng robotics unit raised more than $900 million at post-money valuation above $6.3 billion (company: largest single private financing in China's embodied AI). IDG led; Gaorong; Tencent and Alibaba strategic. He Xiaopeng and Brian Gu invested ~$100M. Iron: 76 DoF body, 21 per hand, three Turing chips 2,250 TOPS combined; mass production expected end-2026, deliveries 2027. Chery AiMOGA reportedly preparing IPO. Hyundai plans Atlas at Georgia factory this year and parts sequencing by 2028; DeepMind partnership; US Robot Metaplant Application Center. Mobileye acquired Mentee Robotics earlier this year for $900 million. Piece notes Tesla has never published an Optimus production count. | ||
| 7:01 PM ET | OPOpenAI | 7:01 PM ETOPOpenAI @LinaOrozcoDus on X OpenAI is killing the official DALL-E GPT today. The o3 model was sunset Wednesday after just 90 days. Google’s gemini-robotics-er shuts down tomorrow. The hidden risk of the AI era is Silent API Breakage. Model lifecycles are shrinking from years down to weeks. If your engineering team is hardcoding specific model IDs into your app instead of using a dynamic fallback routing gateway, you are literally days away from a production blackout. How are you handling high-frequency model deprecation in your CI/CD pipelines? | |
| 10:05 AM ET | VIVisiveAI | 10:05 AM ETVIVisiveAI VisiveAI on X: "Teamwork makes the dream work. To test multi-robot collaboration with Gemini Robotics 2, we challenged our Apollo and Duo robo...
via @GoogleDeepMind
#AI #LLM #MachineLearning #ChatGPT #GenerativeAI" / X Post Log in Sign up Post VisiveAI on X: "Teamwork makes the dream work. To test multi-robot collaboration with Gemini Robotics 2, we challenged our Apollo and Duo robo... via @GoogleDeepMind \#AI \#LLM \#MachineLearning \#ChatGPT \#GenerativeAI" - VisiveAI @VisiveAI Teamwork makes the dream work. To test multi-robot collaboration with Gemini Robotics 2, we challenged our Apollo and Duo robo... via @GoogleDeepMind #AI #LLM #MachineLearning #ChatGPT #GenerativeAI 00:00 View media 10:05 AM · Aug 30, 202646Views Reply Scan to get the app | |
| 7:01 AM ET | SASanctuary AI | 7:01 AM ETSASanctuary AI Sanctuary AI puts humanoid tech on industrial robots Skip to content Sanctuary AI has deployed technology developed alongside its Phoenix humanoid on an industrial wire insertion task, reporting a success rate above 99.5% and a 2.54 second cycle time. The trial used robotic hands rather than a complete humanoid to insert flexible wired plugs into moving targets on a live conveyor at a global Tier 1 automotive supplier. The result came from 313 insertion trials conducted over 40 minutes, CEO Daniel Friedmann told Forbes. Sanctuary calculated the success rate by comparing successful and unsuccessful insertions. Friedmann said the cycle time was… | |
| 7:00 AM ET | $GOOGL | 7:00 AM ET METAL LAB on X: "Google DeepMind posted a video of humanoid Apollo, running Gemini Robotics 2, on Aug 7.
Apollo said tying a trash bag was harder than a full-body balance routine.
It also worked with robot Duo to close and store a garage bin. @GoogleDeepMind" / X Post Log in Sign up Post METAL LAB on X: "Google DeepMind posted a video of humanoid Apollo, running Gemini Robotics 2, on Aug 7. Apollo said tying a trash bag was harder than a full-body balance routine. It also worked with robot Duo to close and store a garage bin. @GoogleDeepMind" - METAL LAB @metallabai Google DeepMind posted a video of humanoid Apollo, running Gemini Robotics 2, on Aug 7. Apollo said tying a trash bag was harder than a full-body balance routine. It also worked with robot Duo to close and store a garage bin. @GoogleDeepMind 00:00 View media 7:00 AM · Aug 30, 202622Views 1… | |
| 11:00 PM ET | $GOOGL | 11:00 PM ET Gemini Robotics 2 brings whole body intelligence to robots - YouTube Error 401 (Bad Request)!!1 401. That’s an error. The server cannot process the request because it is malformed. It should not be retried. That’s all we know. Back Skip navigation Search Search with your voice Sign in Gemini Robotics 2 brings whole body intelligence to robots Tap to unmute 2x Gemini Robotics 2 brings whole body intelligence to robots Google DeepMind 324,052 views 1 month ago Copy link Info Shopping If playback doesn't begin shortly, try restarting your device. • You're signed out Videos you watch may be added to the TV's watch history and influence TV recommendations. To avoid… | |
| 4:12 PM ET | $GOOGL | 4:12 PM ET @csentropy on X Announced July 30, 2026 (author Carolina Parada) as part of Gemini Robotics 2, three models: Gemini Robotics 2 (VLA) converts vision+language into motor control first DeepMind VLA to control a full humanoid feet-to-fingertips under one checkpoint. Partner-only. Gemini Robotics ER 2 (VLM) the embodied-reasoning brain; publicly available. Gemini Robotics On-Device 2 (VLA) runs locally, adapts to new embodiments in a few hours. Partner-only. | |
| 10:05 AM ET | HYHyundai | 10:05 AM ETHYHyundai @StragglerLiu on X At Hyundai Motor's 2026 CEO Investor Day on August 26, CEO José Muñoz made a statement that deserves serious attention: Hyundai is considering selling Boston Dynamics' humanoid robot Atlas through its existing automotive dealer network. Hyundai Capital is exploring financing options for robot sales — the same structures used to finance car purchases. This is not another "automaker experiments with robotics" story. This is Hyundai putting a humanoid robot into the commercial infrastructure it spent 60 years building to sell cars. The roadmap: 2026: RMAC (Robot Metaplant Application Center) in… | |
| 10:00 AM ET | $GOOGL | 10:00 AM ET Piloting the world's first double-blind AI evaluations — Google DeepMind Skip to main content August 27, 2026 Responsibility & Safety Piloting the world's first double-blind AI evaluations William Isaac, Sol Messing and Kristian Lum Share Building trust in proprietary model benchmarks using cryptographically secure environments Imagine a student is set to take a high-stakes exam. If they accidentally peek at the test questions in advance, achieving a perfect score is influenced by this knowledge, making it a meaningless accomplishment. To truly measure what they know, they must have no visibility of the test questions until it's time to take the exam. That is the… | |
| 1:24 PM ET | 1:24 PM ET Humanoids are for marketing, robots are the real business READ00% The Apollo that Apptronik showed the world at the end of June has two legs, a torso, a head and the general proportions of a five-foot-eight adult. The Apollo its paying customers keep asking for has a wheeled base instead. Apptronik, an Austin robotics company that has raised nearly a billion dollars to build a humanoid, now ships its flagship in two configurations, and the one destined for warehouses rolls. Logistics buyers, it turns out, want stability, battery life and predictability more than they want a machine that walks. The case for legs is a good one, and the company has… | ||
| 6:10 AM ET | $GOOGL | 6:10 AM ET @zhodonx on X I’ve said before that I think Google & DeepMind are playing a longer game with models. This is still a firm stance. Because a lot of the time, i think people look at Google’s model strategy too narrowly; Every Gemini release immediately gets reduced to how it compares with other labs or whatever benchmark is trending that week. Meanwhile the it’s more about where all of that intelligence might show up next. Like in Reasoning, Vision, Video, Agents. And particularly robotics. I believe Gemini Robotics ER 2 is a clear example here. It was built on Gemini 3.5 Flash. Except, now its intelligence… | |
| 12:00 PM ET | $GOOGL | 12:00 PM ET Gemini Robotics 2 brings whole body intelligence to robots — Google DeepMind Skip to main content July 30, 2026 Models Gemini Robotics 2 brings whole body intelligence to robots Carolina Parada Share Your browser does not support the video tag. From feet to fingertips — we are teaching robots intelligent whole-body control, fine dexterity, and teamwork to complete a broad range of complex tasks For decades, we’ve dreamed of robots that can seamlessly step into our world and lend a hand. Now, that vision takes a significant stride forward. Most robots are pre-programmed or teleoperated for narrow, repetitive task sequences. They lack the ability to truly learn for… | |
| 12:00 PM ET | $GOOGL | 12:00 PM ET Gemini Robotics 2 brings whole body intelligence to robots — Google DeepMind Skip to main content July 30, 2026 Models Gemini Robotics 2 brings whole body intelligence to robots Carolina Parada Share Your browser does not support the video tag. From feet to fingertips — we are teaching robots intelligent whole-body control, fine dexterity, and teamwork to complete a broad range of complex tasks For decades, we’ve dreamed of robots that can seamlessly step into our world and lend a hand. Now, that vision takes a significant stride forward. Most robots are pre-programmed or teleoperated for narrow, repetitive task sequences. They lack the ability to truly learn for… | |
| — | SUStanford University | —SUStanford University OpenVLA: OpenVLA: An Open-Source Vision-Language-Action Model Moo Jin Kim∗,1 Karl Pertsch∗,1,2 Siddharth Karamcheti∗,1,3 Ted Xiao4 Ashwin Balakrishna3 Suraj Nair3 Rafael Rafailov1 Ethan Foster1 Grace Lam Pannag Sanketi4 Quan Vuong5,† Thomas Kollar3 Benjamin Burchfiel3 Russ Tedrake3,6 Dorsa Sadigh1 Sergey Levine2 Percy Liang1 Chelsea Finn1 https://openvla.github.io970k Robot Episodes ViT Llama 2 7B Base VLM OpenVLA Vision-Language-Action Model Fine-tune VLM w/ Robot Actions: Closed-Loop Robot Control Policy User: Wipe the table. OpenVLA: [ x, , Grip] = …Δ Δθ Δ Multi-Robot Control & Efficient… |
| Time | Entity | Sector | Event / capture |
|---|---|---|---|
| 4:28 PM ET | ANAnthropic | 4:28 PM ETANAnthropic @MParekh on X ‘Nothing to See Here’ in AI, Says DC. Anthropic, OpenAI & Google. ARD #155 ...Commerce clears Anthropic. DOJ backs OpenAI on fair use. Google keeps its ad stack. Ternus up next at Apple Sep 9. AI Ramblings Daily podcast #155 on AI-RTZ. $GOOG Detailed writeup and source links: https://t.co/FRsrX7vCMd | |
| 4:51 AM ET | $GOOGL | 4:51 AM ET @tchsignal on X Gemini 3.8 Flash Puts Google's Rapid AI Release Cycle to the Test Google has released Gemini 3.8 Flash, its third Flash release in six weeks and roughly three weeks after Gemini 3.7 Flash. The new model is generally available with a 1 million-token context window, up to 64,000 output tokens, and low, medium, or high thinking levels. For developers, the important detail is not just Google's rapid release pace. Gemini 3.8 Flash keeps the same introductory per-token pricing as 3.7 Flash: $0.75 per million input tokens and $3.75 per million output tokens through December 31, 2026. But equal token… | |
| 7:54 PM ET | OPOpenAI | — | 7:54 PM ETOPOpenAI @stretchcloud on X The open-source alternative to Claude Managed Agents just launched. Omnara handles the production scaffolding your agent needs: durable execution, sandboxes, permissions, secrets, streaming, approvals.The insight in the positioning is right: most people building agents want to own the agent logic. They don't want the hosting burden. Claude Managed Agents (and equivalents from OpenAI, Google) bundle those two concerns together. Omnara unbundles them.The category is moving fast. Alongside Omnara there's OpenMA, open-managed-agents, and TrueForge from TrueFoundry doing similar things. The… |
| 6:34 PM ET | $GOOGL | — | 6:34 PM ET @stretchcloud on X My read on Muse Spark 1.3: "almost too cheap to meter" is doing real work as a positioning statement.The benchmark picture: 75.4 on DeepSWE v1.1, 88.8 on Terminal-Bench 2.1, 98.5 on MRCR 256K-512K long-context. That last number matters more than the coding scores right now. Most agentic coding tasks need to hold a full repo in working context. Muse Spark leads the long-context category against Opus 5 and GPT-5.6 Sol.The 20% tool call reduction and 25% token usage reduction vs Muse Spark 1.2 compounds fast across multi-step agent plans. Google Agentspace, AWS Bedrock Agents, LangChain, and the… |
| 5:13 PM ET | ANAnthropic | 5:13 PM ETANAnthropic @MParekh on X ‘New and Improved’ in AI Models, Anthropic, OpenAI & Google. ARD #154 ...Fable 5.1’s fine print. Astra tips out on cyber. Gemini narrows the gap. World models level up. Full writeup and source links: https://t.co/UZSXoRFmE2 https://t.co/UOEqsFebFq | |
| 4:20 PM ET | $GOOGL | 4:20 PM ET @stretchcloud on X The model tier split just happened. Gemini 3.8 Flash and Gemini 3.8 Flash Cyber launched today. Same day, completely different access regimes. The public Flash model is on the Gemini API at $0.75 per million input tokens. It scores 73.7% on DeepSWE 1.1, leads Harvey's Legal Agent Benchmark, and is the third updated Flash in 6 weeks. Gemini 3.8 Flash Cyber is not on the public API at all. Access is through Google's Fairwind Program, restricted to government agencies, critical infrastructure operators, and software maintainers. The capability gap is not marginal. Google's Chrome Security team… | |
| 11:11 PM ET | — | 11:11 PM ET @Avin_lucia on X @Arifx0001 @BitRobotNetwork Great synergy leveraging high-end and affordable tools for teleop to scale AI efficiently | |
| 10:38 PM ET | $GOOGL | 10:38 PM ET @OdinsDeposition on X @dotkrueger They essentially pinpointed their long running targets today with Nvidia and Spacex CEOs both aligned on listing 10x economic gains from robotics. Prior to that 20-30% global economic gain from widespread digital AI. | |
| 1:29 PM ET | $GOOGL | 1:29 PM ET Google DeepMind on X: "We’re bringing agentic video understanding to our latest Gemini models.
They can now analyze videos with better accuracy while using up to 88% fewer tokens. 🧵" / X Post Log in Sign up Post Google DeepMind on X: "We’re bringing agentic video understanding to our latest Gemini models. They can now analyze videos with better accuracy while using up to 88% fewer tokens. 🧵" - Google DeepMind @GoogleDeepMind We’re bringing agentic video understanding to our latest Gemini models. They can now analyze videos with better accuracy while using up to 88% fewer tokens. 🧵 View media 1:29 PM · Sep 1, 2026127.2KViews 93 132 1.3K 264 - Google DeepMind @GoogleDeepMind 21h Instead of scanning an entire file, Gemini reasons across the video’s transcript, audio, and… | |
| 10:31 PM ET | $GOOGL | 10:31 PM ET smartfitguide on X: "Google DeepMind's RT-2 model treats robot actions as text tokens. By co-fine-tuning pre-trained Vision-Language Models with robotic data, it achieved nearly double the performance on novel tasks compared to its predecessor, RT-1. #AI" / X Post Log in Sign up Post smartfitguide on X: "Google DeepMind's RT-2 model treats robot actions as text tokens. By co-fine-tuning pre-trained Vision-Language Models with robotic data, it achieved nearly double the performance on novel tasks compared to its predecessor, RT-1. \#AI" - smartfitguide @bareani21645 Google DeepMind's RT-2 model treats robot actions as text tokens. By co-fine-tuning pre-trained Vision-Language Models with robotic data, it achieved nearly double the performance on novel tasks compared to its predecessor, RT-1. #AI View media 10:31 PM · Aug 31, 20268Views Reply 1 | |
| 9:50 PM ET | $GOOGL | 9:50 PM ET 大仏ヨヨ on X: "Gemini Robotics ER 2を試してみたけど結構いいね。リアルタイムでOpenDuckMiniを追跡させてみたけどレスポンスは想定していたより悪くはなかった。どういうロボットかまで判別できているのが良かった。これ以外の機能もあるのでもう少し深堀しようと思う。
#OpenDuckMini,#microduckと同じ作者" / X Post Log in Sign up Post 大仏ヨヨ on X: "Gemini Robotics ER 2を試してみたけど結構いいね。リアルタイムでOpenDuckMiniを追跡させてみたけどレスポンスは想定していたより悪くはなかった。どういうロボットかまで判別できているのが良かった。これ以外の機能もあるのでもう少し深堀しようと思う。 \#OpenDuckMini,\#microduckと同じ作者" - 大仏ヨヨ @U6kcXA6yRN47345 Gemini Robotics ER 2を試してみたけど結構いいね。リアルタイムでOpenDuckMiniを追跡させてみたけどレスポンスは想定していたより悪くはなかった。どういうロボットかまで判別できているのが良かった。これ以外の機能もあるのでもう少し深堀しようと思う。 #OpenDuckMini, #microduckと同じ作者 00:00 View media 9:50 PM · Aug 31, 2026721Views Reply 1 12 4 | |
| 8:00 PM ET | $GOOGL | 8:00 PM ET Introducing Agentic Video in Gemini Introducing agentic video understanding with Gemini Sep 01, 2026 \| 7 min read - x.com - Facebook - LinkedIn - Mail - Copy link Our new agentic feature for video analysis cuts token consumption by up to 88%, reduces costs by up to 66%, and boosts quality by up to 7%. Rohan Doshi Senior Product Manager, Google DeepMind Mario Lučić Research Director, Google DeepMind Share - x.com - Facebook - LinkedIn - Mail - Copy link Your browser does not support the audio element. Listen to article \[\[duration\]\] minutes This content is generated by Google AI. Generative AI is experimental… | |
| 3:11 AM ET | $TSLA | 3:11 AM ET Humanoid Robots News | AI Breakthroughs, Robotics Trends & Synthetic Updates Humanoid Press - Humanoid News - Humanoid Daily - News Archives - Humanoid TV - History Board - AI News - 2026 - Top 10 Artificial Intelligence - About AI - About AGI - Workshop - Operation Instructions - Humanoid & AI Safety - Quantum Robotics - Database - Supplier Directory - CubeMars - Shopping Guide - HPS Leadership - HPS Field Test Audit Checklist - Top 10 Humanoid Company 2026 - Top 10 Humanoid What Each Does Best 2026 - Partnerships - Interview - Opinion - Magazine - About - Copyright Notice - Cookie Policy - Disclaimer - AI & Automation - Licensing & Attribution Policy - Terms of Use… | |
| 8:00 PM ET | $GOOGL | 8:00 PM ET Gemini Robotics ER 1.6 shutdown date Aug 31, 2026; replacement ER 2 Google Developers Gemini API deprecations page, Robotics models table (page last updated 2026-08-27 UTC; shutdown date in the table is in-window): gemini-robotics-er-1.6-preview released April 14, 2026, shutdown date August 31, 2026, recommended replacement gemini-robotics-er-2-preview. Older gemini-robotics-er-1.5-preview listed shutdown April 30, 2026. Companion overview (ai.google.dev/gemini-api/docs/robotics-overview, scraped Sep 1) says upgrade by replacing model='gemini-robotics-er-1.6-preview' with gemini-robotics-er-2-preview or gemini-robotics-er-2-streaming-preview; ER 1.6 'will be shut down at the end of August.' Pricing page (same scrape): ER 2 Preview paid $2.00 input / $10.00 output per 1M tokens (text/image/video/audio); batch $1.00 / $5.00; ER 2 Streaming $2.00 / $10.00; ER 1.6 Preview $1.00 input ($2.00 audio) / $5.00 output. This is an API lifecycle event, not a factory-hours print. Gemini Robotics 2 VLA launch itself was July 30, 2026 (Hassabis) — outside. | |
| 8:00 PM ET | 8:00 PM ET Apptronik Forge tape Aug 31: $36.97 (-7.94%), $5B mark; $964.7M total funding Forge Global Apptronik private-share page (Forge Price updated Aug 31, 2026): Forge Price $36.97, -$3.19 (-7.94%); Forge Price valuation $5B. Total funding listed $964.7M. Funding table: Series A-X1 02/11/2026 $441.21M at $36.97/share, $5B post-money; Series A-X2 02/11/2026 $78.79M at $36.97, $5B post-money; investors named AT&T Ventures, B Capital, Google, John Deere, Mercedes-Benz Group, Peak6, Qatar Investment Authority. Earlier Series A-1/A-2/A-3 03/18/2025 totaling about $415M at $1.78B post-money. Active-market widget timestamped 8/27/2026 2:33:46 PM EDT (3 bids, total bids $215.95K). News module still points at Robot Park coverage dated 06/30/2026 Forbes and 07/06/2026 Robotics & Automation News — those factory items are outside this window; the in-window fact is the Aug 31 private mark and listed capitalization, not a new Robot Park opening. | ||
| 8:00 PM ET | 8:00 PM ET Xpeng robotics unit: >$900M at >$6.3B; Iron mass production targeted end-2026 Vytautas Valinskas / Technology.org (Aug 31, 2026), largely recapping TechCrunch Aug 28: Xpeng robotics unit raised more than $900 million at post-money valuation above $6.3 billion (company: largest single private financing in China's embodied AI). IDG led; Gaorong; Tencent and Alibaba strategic. He Xiaopeng and Brian Gu invested ~$100M. Iron: 76 DoF body, 21 per hand, three Turing chips 2,250 TOPS combined; mass production expected end-2026, deliveries 2027. Chery AiMOGA reportedly preparing IPO. Hyundai plans Atlas at Georgia factory this year and parts sequencing by 2028; DeepMind partnership; US Robot Metaplant Application Center. Mobileye acquired Mentee Robotics earlier this year for $900 million. Piece notes Tesla has never published an Optimus production count. | ||
| 7:01 PM ET | OPOpenAI | 7:01 PM ETOPOpenAI @LinaOrozcoDus on X OpenAI is killing the official DALL-E GPT today. The o3 model was sunset Wednesday after just 90 days. Google’s gemini-robotics-er shuts down tomorrow. The hidden risk of the AI era is Silent API Breakage. Model lifecycles are shrinking from years down to weeks. If your engineering team is hardcoding specific model IDs into your app instead of using a dynamic fallback routing gateway, you are literally days away from a production blackout. How are you handling high-frequency model deprecation in your CI/CD pipelines? | |
| 10:05 AM ET | VIVisiveAI | 10:05 AM ETVIVisiveAI VisiveAI on X: "Teamwork makes the dream work. To test multi-robot collaboration with Gemini Robotics 2, we challenged our Apollo and Duo robo...
via @GoogleDeepMind
#AI #LLM #MachineLearning #ChatGPT #GenerativeAI" / X Post Log in Sign up Post VisiveAI on X: "Teamwork makes the dream work. To test multi-robot collaboration with Gemini Robotics 2, we challenged our Apollo and Duo robo... via @GoogleDeepMind \#AI \#LLM \#MachineLearning \#ChatGPT \#GenerativeAI" - VisiveAI @VisiveAI Teamwork makes the dream work. To test multi-robot collaboration with Gemini Robotics 2, we challenged our Apollo and Duo robo... via @GoogleDeepMind #AI #LLM #MachineLearning #ChatGPT #GenerativeAI 00:00 View media 10:05 AM · Aug 30, 202646Views Reply Scan to get the app | |
| 7:01 AM ET | SASanctuary AI | 7:01 AM ETSASanctuary AI Sanctuary AI puts humanoid tech on industrial robots Skip to content Sanctuary AI has deployed technology developed alongside its Phoenix humanoid on an industrial wire insertion task, reporting a success rate above 99.5% and a 2.54 second cycle time. The trial used robotic hands rather than a complete humanoid to insert flexible wired plugs into moving targets on a live conveyor at a global Tier 1 automotive supplier. The result came from 313 insertion trials conducted over 40 minutes, CEO Daniel Friedmann told Forbes. Sanctuary calculated the success rate by comparing successful and unsuccessful insertions. Friedmann said the cycle time was… | |
| 7:00 AM ET | $GOOGL | 7:00 AM ET METAL LAB on X: "Google DeepMind posted a video of humanoid Apollo, running Gemini Robotics 2, on Aug 7.
Apollo said tying a trash bag was harder than a full-body balance routine.
It also worked with robot Duo to close and store a garage bin. @GoogleDeepMind" / X Post Log in Sign up Post METAL LAB on X: "Google DeepMind posted a video of humanoid Apollo, running Gemini Robotics 2, on Aug 7. Apollo said tying a trash bag was harder than a full-body balance routine. It also worked with robot Duo to close and store a garage bin. @GoogleDeepMind" - METAL LAB @metallabai Google DeepMind posted a video of humanoid Apollo, running Gemini Robotics 2, on Aug 7. Apollo said tying a trash bag was harder than a full-body balance routine. It also worked with robot Duo to close and store a garage bin. @GoogleDeepMind 00:00 View media 7:00 AM · Aug 30, 202622Views 1… | |
| 11:00 PM ET | $GOOGL | 11:00 PM ET Gemini Robotics 2 brings whole body intelligence to robots - YouTube Error 401 (Bad Request)!!1 401. That’s an error. The server cannot process the request because it is malformed. It should not be retried. That’s all we know. Back Skip navigation Search Search with your voice Sign in Gemini Robotics 2 brings whole body intelligence to robots Tap to unmute 2x Gemini Robotics 2 brings whole body intelligence to robots Google DeepMind 324,052 views 1 month ago Copy link Info Shopping If playback doesn't begin shortly, try restarting your device. • You're signed out Videos you watch may be added to the TV's watch history and influence TV recommendations. To avoid… | |
| 4:12 PM ET | $GOOGL | 4:12 PM ET @csentropy on X Announced July 30, 2026 (author Carolina Parada) as part of Gemini Robotics 2, three models: Gemini Robotics 2 (VLA) converts vision+language into motor control first DeepMind VLA to control a full humanoid feet-to-fingertips under one checkpoint. Partner-only. Gemini Robotics ER 2 (VLM) the embodied-reasoning brain; publicly available. Gemini Robotics On-Device 2 (VLA) runs locally, adapts to new embodiments in a few hours. Partner-only. | |
| 10:05 AM ET | HYHyundai | 10:05 AM ETHYHyundai @StragglerLiu on X At Hyundai Motor's 2026 CEO Investor Day on August 26, CEO José Muñoz made a statement that deserves serious attention: Hyundai is considering selling Boston Dynamics' humanoid robot Atlas through its existing automotive dealer network. Hyundai Capital is exploring financing options for robot sales — the same structures used to finance car purchases. This is not another "automaker experiments with robotics" story. This is Hyundai putting a humanoid robot into the commercial infrastructure it spent 60 years building to sell cars. The roadmap: 2026: RMAC (Robot Metaplant Application Center) in… | |
| 10:00 AM ET | $GOOGL | 10:00 AM ET Piloting the world's first double-blind AI evaluations — Google DeepMind Skip to main content August 27, 2026 Responsibility & Safety Piloting the world's first double-blind AI evaluations William Isaac, Sol Messing and Kristian Lum Share Building trust in proprietary model benchmarks using cryptographically secure environments Imagine a student is set to take a high-stakes exam. If they accidentally peek at the test questions in advance, achieving a perfect score is influenced by this knowledge, making it a meaningless accomplishment. To truly measure what they know, they must have no visibility of the test questions until it's time to take the exam. That is the… | |
| 1:24 PM ET | 1:24 PM ET Humanoids are for marketing, robots are the real business READ00% The Apollo that Apptronik showed the world at the end of June has two legs, a torso, a head and the general proportions of a five-foot-eight adult. The Apollo its paying customers keep asking for has a wheeled base instead. Apptronik, an Austin robotics company that has raised nearly a billion dollars to build a humanoid, now ships its flagship in two configurations, and the one destined for warehouses rolls. Logistics buyers, it turns out, want stability, battery life and predictability more than they want a machine that walks. The case for legs is a good one, and the company has… | ||
| 6:10 AM ET | $GOOGL | 6:10 AM ET @zhodonx on X I’ve said before that I think Google & DeepMind are playing a longer game with models. This is still a firm stance. Because a lot of the time, i think people look at Google’s model strategy too narrowly; Every Gemini release immediately gets reduced to how it compares with other labs or whatever benchmark is trending that week. Meanwhile the it’s more about where all of that intelligence might show up next. Like in Reasoning, Vision, Video, Agents. And particularly robotics. I believe Gemini Robotics ER 2 is a clear example here. It was built on Gemini 3.5 Flash. Except, now its intelligence… | |
| 12:00 PM ET | $GOOGL | 12:00 PM ET Gemini Robotics 2 brings whole body intelligence to robots — Google DeepMind Skip to main content July 30, 2026 Models Gemini Robotics 2 brings whole body intelligence to robots Carolina Parada Share Your browser does not support the video tag. From feet to fingertips — we are teaching robots intelligent whole-body control, fine dexterity, and teamwork to complete a broad range of complex tasks For decades, we’ve dreamed of robots that can seamlessly step into our world and lend a hand. Now, that vision takes a significant stride forward. Most robots are pre-programmed or teleoperated for narrow, repetitive task sequences. They lack the ability to truly learn for… | |
| 12:00 PM ET | $GOOGL | 12:00 PM ET Gemini Robotics 2 brings whole body intelligence to robots — Google DeepMind Skip to main content July 30, 2026 Models Gemini Robotics 2 brings whole body intelligence to robots Carolina Parada Share Your browser does not support the video tag. From feet to fingertips — we are teaching robots intelligent whole-body control, fine dexterity, and teamwork to complete a broad range of complex tasks For decades, we’ve dreamed of robots that can seamlessly step into our world and lend a hand. Now, that vision takes a significant stride forward. Most robots are pre-programmed or teleoperated for narrow, repetitive task sequences. They lack the ability to truly learn for… | |
| — | SUStanford University | —SUStanford University OpenVLA: OpenVLA: An Open-Source Vision-Language-Action Model Moo Jin Kim∗,1 Karl Pertsch∗,1,2 Siddharth Karamcheti∗,1,3 Ted Xiao4 Ashwin Balakrishna3 Suraj Nair3 Rafael Rafailov1 Ethan Foster1 Grace Lam Pannag Sanketi4 Quan Vuong5,† Thomas Kollar3 Benjamin Burchfiel3 Russ Tedrake3,6 Dorsa Sadigh1 Sergey Levine2 Percy Liang1 Chelsea Finn1 https://openvla.github.io970k Robot Episodes ViT Llama 2 7B Base VLM OpenVLA Vision-Language-Action Model Fine-tune VLM w/ Robot Actions: Closed-Loop Robot Control Policy User: Wipe the table. OpenVLA: [ x, , Grip] = …Δ Δθ Δ Multi-Robot Control & Efficient… |