NAll news

All news

Friday1456 items

  1. Garry Tan
    AI score15

    Garry Tan mocks school district for cutting honors classes while adding new courses

    Citing a post from Garry's List, Garry Tan pointed out that Palo Alto Unified, while facing a $33 million budget shortfall, is phasing out high school honors courses yet plans to offer new courses. He did not specify what the new courses are, expressing his sarcasm only through a rhetorical question.

    Quoted postGarry's List · @garryslist

    Not to beat a dead horse, but really glad Palo Alto Unified is prioritizing the important stuff during a $33M budget crunch. Which, apparently, isn’t honors classes. https://t.co/7CJ8KMlyum

    #English text is being prepared.

  2. Thariq
    AI score8

    Thariq: Artists Worry About AI, But I Want to Use Agents to Complete Creative Ideas

    Thariq believes a similar phenomenon exists in the creative field: he sees artists deeply worried about AI, but he very much wants to first sketch out animation ideas and character concept drafts, and then let agents refine them, rather than painstakingly searching for reference examples. He laments his own lack of this expressive ability.

    Quoted postThariq · @trq212

    very weird thing happening rn technical people are sad, they think their skills are now useless non-technical people are desperately wishing they were more technical so they could use agents better

    #English text is being prepared.#Agents

  7. Elon Musk
    AI score22

    Musk: Breakthroughs in Every Field Are Accelerating Exponentially

    Musk posted that, as long as one observes carefully enough, breakthroughs in every field are accelerating exponentially, and exponential growth always appears small in its early stages. He cited a post listing biotech advances from the past six months as background.

    Quoted postMgoes (bio/acc 🤖💉) · @m_goes_distance

    this might be my last normal tweet before biotech goes vertical i'm not a whistleblower. I don't work at Lilly, Novo or any of the big labs but there is a massive shift happening in human biology right now. it's sitting in press releases and almost nobody reads them the last 6 months alone: - a once-daily pill roughly doubled median survival in metastatic pancreatic cancer (13.2 vs 6.6 months) - a one-time gene editing infusion cut hereditary angioedema attacks 87%, and 62% of patients stayed attack-free and off all other prevention for 6 months - a one-time gene edit for cholesterol dropped LDL up to 62%, and it held for over a year - a personalized mrna vaccine beat keytruda alone in a 1,137 patient melanoma phase 3 - retatrutide hit ~28% weight loss in a phase 3 - an ai-designed drug entered phase 3 - a pill that regrows hair passed its second phase 3 - an immunotherapy that could let some rectal cancer patients skip surgery, chemo and radiation just got fda approved - a man lived 271 days off dialysis on a pig kidney - human cells were caught passing chromosome-sized chunks of DNA to their neighbors, and it changes how the neighbors behave we are on the verge of an era where the human body is a software problem and the update cycle gets shorter everyday stick around, it's going to get weird bio/acc

    #English text is being prepared.

  9. Elon Musk
    AI score3

    Musk replies to post about 2008 Tesla and SpaceX history

    X user X Freeze posted a recollection that in 2008, when Tesla and SpaceX were on the brink of collapse, Musk split his remaining $30 million to $40 million between the two companies instead of betting on just one. Musk himself replied only with "True," confirming that the account is accurate.

    Quoted postX Freeze · @XFreeze

    In 2008, Elon Musk faced a decision that would have broken most people Two companies he had poured his heart, soul, and everything he had into were on the brink of collapse. He had just $30 to $40 million left, and saving one meant risking the life of the other He was forced to confront an almost impossible question: Which of the two dreams he had sacrificed so much to build was he willing to let die? His answer was neither Even if it meant risking everything he had left Tesla and SpaceX were both fighting for survival, and he had to make one of the hardest choices of his life "I had two choices. I could put it all into one company, and then the other company would definitely die. Or split it between the two companies. But if I split it between the two companies, then both might die" Then Elon said something that tells you everything about how much these companies meant to him: "When you put your blood, sweat, and tears into creating something or building something, it's like a child. And so it's like which one am I gonna let starve to death? I couldn't bring myself to do it" He split the money between both companies, knowing he was risking losing everything "Fortunately, thank goodness, they both came through" Nearly two decades later, Tesla has transformed the electric vehicle industry, and SpaceX has revolutionized spaceflight Two companies that were on the brink of collapse went on to change the world All because Elon Musk couldn't bring himself to let either one starve

    #English text is being prepared.

  11. Rohan Paul
    AI score60

    ServiceNow Beats Earnings Expectations, Yet Stock Still Falls 19% After Hours

    In the September quarter, ServiceNow's revenue grew 20.5% year over year to $4.106 billion, exceeding its own quarterly target, but the stock still fell 19% after hours as the market worried that AI agents are eroding its per-seat SaaS business. Signed revenue for the next 12 months grew 20.5% year over year to $13.65 billion, and the company expects that growth rate to slow to 18.5% this quarter, while total revenue guidance is about $4.25 billion, up about 23%. AI transaction volume grew 80% year over year, and the 2026 AI annual contract value target was raised from $1 billion in January to $1.7 billion, but it remains small compared with the expected subscription revenue of about $15.77 billion in 2026.

    Same story · 2 reports

    #Industry news#Agents#English text is being prepared.

  12. Rohan PaulResearch
    AI score62

    Google Releases EmbeddingGemma 2 Paper: A 744M-Parameter On-Device Multimodal Embedding Model

    Google has released the EmbeddingGemma 2 paper, an Apache 2.0-licensed on-device multimodal embedding model that maps text, code, images, audio, and video into the same vector space. The model is built on Gemma 4 and tuned through contrastive training by learning from the larger Gemini Embedding 2; with 744M parameters, it surpasses models 2 to 12 times larger on multiple tasks. On the Pixel 11 Pro, the quantized text weights occupy about 191MB of active memory, while the full multimodal model takes about 567MB; the context window is expanded 4 times to 8K tokens, accommodating about 5.5 minutes of audio, 29 images, or 58 video frames; directly embedding speech scores 69.54 on the MSEB spoken query retrieval benchmark, higher than the 57.59 of transcribing first and then retrieving, and code retrieval on MTEB Code improves from 68.76 in EmbeddingGemma 1 to 78.68.

    #Papers/Research#Multimodal#English text is being prepared.

  13. Rohan PaulResearch
    AI score50

    UC Santa Barbara paper proposes TTME: replacing test-time training with evolutionary memory for mathematical discovery

    A UC Santa Barbara paper proposes TTME, which replaces test-time training with evolutionary memory in mathematical discovery tasks, keeping weights unchanged and storing rules and score-improvement ideas from previous attempts in memory to guide subsequent attempts. Across 5 mathematical tasks, 2 frameworks, and 3 models, TTME exceeded the reported test-time training scores every time; on 3 tasks with Qwen3-8B, its FLOPs were reduced by 82.8% to 84.6%, and runtime was reduced by 47.7% to 58.4%. The authors suggest that if an LLM search loop comes with an automatic checker, one can first try using experiential memory rather than paying the cost of weight updates in the middle of the search.

    #Papers/Research#Reasoning#English text is being prepared.

  14. Rohan Paul
    AI score61

    Microsoft returns to the US bond market, planning to issue at least $25 billion in bonds

    Bloomberg reports that Microsoft is returning to the US bond market for the first time since its $17 billion bond issuance in 2017, planning to raise at least $25 billion through up to 10 tranches. Its 40-year bond pricing is close to the long-term bonds of Alphabet and Amazon, despite Microsoft having a higher AAA rating. Hyperscale cloud providers and data center developers have issued more than $313 billion in investment-grade bonds this year, nearly three times the total for 2025, driving global bond issuance to a record $6.61 trillion.

    Same story · 2 reports

    #Industry news#OpenAI

  15. Rohan Paul
    AI score67

    Oracle plans to increase computing capacity to approximately 10GW within three years, with AI remaining performance obligations rising to $638B

    At its Investor Day, Oracle laid out its computing expansion roadmap, increasing from approximately 1GW in FY25 to approximately 3GW in FY26, and plans to reach approximately 10GW by FY28. Its remaining performance obligations (RPO) increased from $138B at the end of FY25 to $638B at the end of FY26, adding approximately $500B in contracted business in one year. Based on Oracle's own estimate of approximately $11B to $16B in annual revenue per 1GW at full capacity, approximately 7GW corresponds to annual revenue of approximately $77B to $112B.

    #Industry news#English text is being prepared.#OpenAI

  16. Rohan PaulProduct release
    AI score62

    Snaptrude launches a building design Agent that can generate LOD 400 construction-level models directly from a site

    Snaptrude has launched a building design Agent that can start from a site or a brief to generate LOD 400 construction-level models including structure, MEP and complete drawings, and supports exporting back to existing software while preserving component families. During early access, more than 2,000 designers have run over 25,000 tasks, and the company has also released the AECBench benchmark along with its first batch of results. Rohan Paul relayed that most architectural AI stops at the level of form, whereas this model carries actual components that can be directly manufactured and installed.

    Quoted postAltaf Ganihar · @altafganihar

    Today we're launching @snaptrude Agent, the most advanced agent for building design. It does real design work with you, from a site to a construction-ready LOD 400 model with structure, MEP and drawings. Then can go back to your existing software with your families intact. About a year ago we bet the company on AI. Honestly the insight came from software engineering. It wasn't long ago that autocomplete felt like a big deal, and now engineers direct agents that write a lot of the code. I think the same shift is coming to how we design buildings, and it's already started. Most designers I meet aren't short on skill, they're short on time. Take the call every design team dreads, the client cuts the budget by 30%. Today that's a bad month for everyone. With the agent you're the director. You decide where the cost comes out, and it updates the layout, coordinates the structure and MEP, and redoes the drawings. So when design exploration gets this cheap, we go from 3 options to 300, and more options means better buildings. Since August our users have gone through more than 140 billion tokens, and firms like Jacobs, Studio+, Kajima, VMDO, Dekker and hundreds of others use Snaptrude today. In our early access, more than 2,000 designers have run over 25,000 tasks through the agent since September. We also built a benchmark for this, AECBench, because in AEC an almost correct answer is still wrong. It gives AI the same brief an architect would get, has its design in a real BIM model, and scores the result against the brief and against how architects review design. We're publishing the first results today, link below. Thank you @prashanthp at @Accel and Shub and the team at @foundamental for backing us when this was still a bet, and the Snaptrude team, who built every bit of this. Really proud of you all. Thanks to our investors, @scottbelsky, @andreasklinger, @thilokonzok, @immad, @CindyXBi, @DanilaVladi, @jmwass, rahulchaudhary, @adamecourt, Zach Ginsburg, Alex duff, Paul grossinger, really appreciate all the help along the way.

    Same story · 2 reports

    #Product updates#Agents#Multimodal

  17. indigo
    AI score31

    Google releases Gemini 4 Carbon; shares fall 10% after hours following earnings report

    Google released Gemini 4 Carbon, less than a month after the previous version, Argon. Shares plunged 10% after hours following the earnings report, mainly because search business growth fell short of expectations and slowed for two consecutive quarters; of total revenue of $127.577 billion, advertising accounted for 64.28% and Google Cloud for 24.07%, with the pure search business falling below 50% for the first time.

    Same story · 34 reports

    #Industry news#Google

  18. Shao MengAnalysis
    AI score59

    DeepLearning.AI Launches Context Engineering Course, Teaching Agent Context Management

    DeepLearning.AI has launched a new Context Engineering course, taught by Chroma CEO Jeff Huber and CMPND CEO Drew Breunig, framing context engineering as the main thread for keeping Agents reliable and token-efficient in complex multi-step tasks. The course builds a harness from scratch in Python and uses traces and evals to quantify each change, covering three capabilities—Agentic Search, Context Management, and Agentic Memory—across four modules, with labs mostly consisting of about 1-hour coding exercises. The course assumes intermediate Python, familiarity with LLM applications and tool calling, and basic retrieval.

    #Tutorials / practice#Agents#RAG

  19. Shao MengResearch
    AI score58

    Google Research releases SLEEP: using offline sleep cycles to mitigate catastrophic forgetting in large models

    Google Research releases the SLEEP method, which lets LLMs periodically switch between wakeful interaction and sleep consolidation, writing short-term contextual information into scalable long-term parameters, and self-improving with synthetic data during the dreaming phase. The sleep phase unlocks reserved MoE memory blocks, combining generalized knowledge distillation and imitation learning to complete memory consolidation, then resets high-frequency temporary parameters. In evaluations of continual translation on unseen languages, SLEEP shows near-zero performance drop; on BABILong it is claimed to scale to about 10 million tokens; on AIME 24/25 and HMMT 25 it achieves approximately 79.2%, 69.0%, and 46.1%, respectively.

    Quoted postGoogle Research · @GoogleResearch

    Today on the blog, learn about our new instance of a “sleep” phase in continual learning, which enables LLMs to continuously integrate new facts and enhance reasoning while mitigating catastrophic forgetting or static knowledge cutoffs. More →https://t.co/5v9w5Pn4dw https://t.co/7Mms1x7hgl

    Same story · 3 reports

    #Papers/Research#English text is being prepared.#Google

  20. Shao MengModel releases
    AI score74

    Google DeepMind Previews Gemini 4 Carbon, Claiming Leadership in Cybersecurity and Long-Horizon Capabilities

    Google DeepMind has previewed its next-generation frontier model, Gemini 4 Carbon, claiming it achieves SOTA in cybersecurity and long-horizon capabilities, ranking first on the Artificial Analysis Intelligence Index v5.0 with 59 points, while Claude Opus 5.5 scored 58. The model is not yet publicly available; it will first be opened to select cybersecurity defenders and trusted partners in the Fairwind program, and will participate in the U.S. government's voluntary pre-release testing, before gradually being opened to developers, enterprises, and consumers. Introductory pricing is the same as Argon, at approximately $2 input and $10 output per million tokens, rising to approximately $4 input and $20 output after the introductory period.

    Quoted postGoogle DeepMind · @GoogleDeepMind

    Gemini 4 Carbon is our next frontier model. It’s state-of-the-art across cybersecurity and long-horizon capabilities.

    Same story · 34 reports

    #Model releases#Safety / alignment#Agents