iCreatAI News
Back to main site
Skip to content
#Data / training
Nov 10Tue
  1. Ars Technica · AI36

    Ukraine’s fiber-optic drones get squeezed by AI data center demand

    Fiber-optic drones, which are immune to battlefield jamming because they are physically connected to operators by miles of cable, have become a common weapon in Ukraine, leaving tangled cables across trees and fields. Importing the fiber-optic material from China has grown more costly as drone warfare needs compete with global AI data center demand for the same limited supply, prompting Ukrainian drone makers to adapt. Russia first deployed unjammable fiber-optic drones in 2024 to repel Ukraine's incursion into Kursk Oblast, and Ukraine quickly followed; both countries have relied on Chinese suppliers for drones and components, with finished spools holding enough cable for ranges of dozens of miles and sometimes up to 62 miles (100 kilometers) in extreme cases, according to The Insider.

  2. Hacker News AI58

    Beyond FLOP/s: comparing AI hardware for inference by memory and communication

    The article argues that raw FLOP/s understates inference gains, estimating Nvidia's GB300 delivers about 25× H100's inference capacity per chip on modeled agentic workloads despite only 2.5× the 8-bit FLOP/s, because memory and inter-chip communication matter more. It introduces the Inference Performance Index (IPI), scored in GB300 equivalents (GB300e), which compares simultaneous requests each chip can serve across 50–400 tokens/s/req for four representative models. Reported scores include GB300 NVL72 at 1, B300 HGX at 0.54, MI355X at about 0.55, VR200 at 2.18, and H200 HGX at 0.12.

  3. Google DeepMind58

    Opening the datasets behind our AI weather models

    Google's WeatherNext team released SatWeave, an hourly global mosaic of geostationary satellite imagery across 17 channels blending six satellites, and PARDIG, an AI-based global precipitation dataset at 15-minute and roughly 10-kilometer resolution trained on NASA GPM radar observations. Both datasets, key to the WeatherNext 3 model, are available as live feeds and historical archives on Google Cloud Storage, with PARDIG catalogs also in Google Earth Engine. PARDIG reportedly improves hourly precipitation detection skill over NASA's IMERG-Final by more than 30% on light-to-moderate precipitation.

  4. Hacker News AI62

    Fireworks AI security breach impacts Notion and their customers

    Notion says less than 1% of its workspaces may have been affected by a security incident at vendor Fireworks AI, which may have exposed unattributable Notion AI chat transcripts used for the Insights feature. Notion disconnected from Fireworks, deleted data in its system, halted training runs, and disabled Insights; affected workspace owners can review the data in their workspace. Notion recommends treating the data as exposed and rotating any sensitive credentials pasted into Notion AI during that timeframe.

  5. Hacker News AI78

    Tell HN: Notion AI chat sessions have been exposed via Fireworks AI hack

    A Notion email to affected customers says a security incident at vendor Fireworks AI may have exposed data, and that Notion shut down all connections to Fireworks' systems as a precaution. On November 6 Fireworks notified Notion that malicious actors may have accessed potential customer data, and on November 8 confirmed a new dataset that included unattributable Notion AI chat transcripts used for the Insights feature, without labels identifying the author or workspace and without tool call results. Notion says the Insights feature has been disabled during the investigation and that it no longer serves open-weight models through Fireworks.

    Why it matters: The post shares a vendor-breach notification email, offering a first-hand look at how an AI infrastructure provider incident can reach downstream customer data.

  6. MIT News AI40

    MIT-led team receives funding to develop next-generation alternatives to modern weed killers

    MIT researchers will lead an effort of up to $53 million, awarded by ARPA-H under its STREAM program, to develop next-generation weed killers that help farmers and protect human health. The project, Precision Learning for On-target Weed control (PLOW), will use generative AI to predict chemical compounds or siRNA that disrupt the same plant enzymes targeted by glyphosate and other herbicides, with candidate molecules screened and then evaluated in greenhouse studies. The team also plans to develop living microbial systems that degrade agrochemicals before they reach food, soil, or waterways.

  7. Hacker News AI58

    The Red Queen hypothesis – a new way forward for self-improving AI

    Researchers, with collaborators from NVIDIA and Flower Labs, propose the Red Queen Gödel Machine, a framework in which a self-improving AI agent and its evaluator co-evolve so the evaluation gets harder as the agent improves, avoiding the ceiling imposed by a fixed test. In experiments on scientific paper writing and reviewing and on Olympiad-level proof writing and grading, co-evolved paper writers reached 1.78×–1.86× higher acceptance rates under an agent-as-a-judge panel and co-evolved graders reached 9% higher ground-truth accuracy. A hybrid setup using NVIDIA Nemotron 3 Ultra alongside ChatGPT-5.5 approached ChatGPT's performance on paper reviewing while cutting search-token costs by around 13 times, though the team stresses the work is preliminary.

  8. Hacker News AI71

    CoreWeave bond yield tops 16% as AI credit stress rises

    CoreWeave's 2032 bond with a 9.625% coupon now yields above 16%, and its five-year credit default swaps spiked above 1,289 basis points Thursday before easing to around 1,155, per Deutsche Bank. Credit default swaps have also risen for Nvidia (up 76% in the last month), Broadcom (up 49%) and Amazon (up 44%), though their spreads remain low in absolute terms. Goldman Sachs estimates moratoriums under incoming Democratic governors in Ohio, Texas and Georgia could affect 45% of planned U.S. data center capacity.

  10. Exponential View22

    Madness and mayhem

    Azeem says the past two weeks have been chaotic in tech, citing a correction in AI companies, the market shock of the Democrats' mid-term victory that gave the Nasdaq 100 its worst day since March 2020, and news that Jensen Huang has a relationship with a Chinese spy. He argues this is a flight to quality rather than a collapse: spending keeps rising, AI revenue reached $188bn in the twelve months to October and runs at $345bn a year, and Anthropic's growth falling to 45% a quarter puts it in line with the market. He expects Democrats to make data centers dearer, favoring hyperscalers, Nvidia and the open-weight camp to have their wings clipped, an Anthropic IPO at a sensible price to steady the sector, and neoclouds to be the weak link.

  11. Hacker News AI33

    Why Is AI Slop So Frustrating to Read?

    This piece argues that language models lag at writing because language is harder to optimize with straightforward rewards and less economically appealing short-term than capabilities like coding, and it focuses on workplace "slop" such as Slack messages and PR descriptions that readers cannot simply scroll past. It identifies three recurring faults—hidden characters and actions, vague wording, and empty statements with misplaced emphasis—and reports that AI samples show much higher rates of unclear communication than human writing across articles and coding discussions.

  12. Hacker News AI42

    I use 11.35B AI tokens a month. How many do you use?

    An analysis of 12,000 developers across 200 companies in Q1 2026 found that the median AI coding user consumes about 51 million tokens per month (roughly $52.38), while 90th-percentile users consume about 380 million tokens (roughly $691.14). Higher token usage correlates with more merged pull requests, but cost per PR rises from $0.28 in the lowest usage tier to $89.32 in the highest, and the median developer uses about 7 million tokens per PR versus roughly 69 million for the top decile. The author argues that broad, moderate adoption delivers more value than pushing a small group to extreme usage, and that an "agentic barrier" limits gains from simply increasing token consumption.

  13. Hacker News AI60

    AI chip financing hits a wall

    Investors who rushed to lend for the data-center buildout are now trying to offload loan portfolios at steep discounts, with GPU financings especially under pressure over fears the loans will outlast the technology's useful life, a person told Semafor. Bloomberg reported Friday that Dutch bank ING was looking to dump $500 million of data-center loans, including a $25 million chunk backing Stargate, the OpenAI, SoftBank and Oracle project. Goldman Sachs and Apollo have begun marketing a $40 billion chip financing deal for SpaceX, the largest of its kind, while Ares Management CEO Michael Arougheti said on an earnings call that his firm would pass on such deals because he does not know how to underwrite the decline and depreciation curves on that technology.

  14. Hacker News AI38

    Triton for MTIA: Bridging the Programming Model Gaps for Custom AI Accelerators

    This paper presents the first production-scale application of Triton on a custom ML accelerator, Meta's MTIA-2i, via a new compiler backend, TorchInductor code generation enhancements, and minimal language extensions exposing MTIA-specific architectural features. Triton-MTIA kernels achieve performance competitive with expert-tuned C++ implementations, and manually written plus Inductor-generated Triton kernels were deployed in production across roughly 60 model types, covering 50% of layers and 47% of non-GEMM execution time for these models.

  15. MIT Technology Review · AI38

    The backlash about data centers might really just be about electricity

    Reporting from Data Center Alley in Ashburn, Virginia, the author argues that public opposition to data centers is driven more by electricity than by the culture war framing Big Tech favors, since facilities drawing 54 MW or more strain local grids and force new transmission costs onto utilities. Because AI-scale sites ideally need around 6,000 MW, companies are increasingly building behind-the-meter natural-gas turbines and diesel backup in rural areas, which still affects surrounding communities through air pollution and a lack of local say. The piece notes that industry insiders see their ambitions constrained less by public opposition than by a lack of electricity, and that future sites may go where few people can resist them.

  16. Hacker News AI58

    Fireworks.ai Update on October 2026 security incident

    Fireworks.ai said its investigation into the October 2026 incident, assisted by Palo Alto Networks Unit 42, found impacts to account and billing information and, for a limited set of accounts, to secrets, datasets and incomplete models, with evidence linking the activity to a financially motivated group engaged in "LLMJacking." A third-party cloud provider hosting Fireworks R&D machines was breached on October 23, and on October 24 an unauthorized party used it to reach Fireworks development machines and run a workload that led to additional customer data impact; Fireworks detected the activity the same day and says it has not detected further successful unauthorized access since October 25. Fireworks has implemented hardening measures including credential lockdown, network isolation and tightened data access, and has disabled self-serve dataset uploads and downloads for Batch Inference and dataset-based training while Inference, Training SDK and Nexus remain ZDR by default.

  17. MarkTechPost64

    Baidu’s ERNIE Evolve Automated Research System Beats NanoChat and NanoGPT Records

    Baidu released ERNIE Evolve, an automated research system that selects research directions, edits code and runs experiments against a fixed objective, keeping only changes that pass a hidden evaluator and a separate reviewer. It beat the best public result on 3 of 4 public metrics, including 0.8763 validation BPB on NanoChat AutoResearch, 38.9 s on NanoGPT Speedrun Track 1 and 2,590 steps on Track 3, while scoring 0.819 mean SOL on SOL-ExecBench, behind the 0.827 leaderboard best. On ERNIE-5.5-Flash inference it reported 623.6% of baseline generation speed on Kunlunxin P800, +33.7% deterministic throughput and +14.9% single-request decode speed, though the system itself is not released.

  18. MarkTechPost60

    Google Unveils TPU 8i Max and TPU 8tx: A Dual-Chip TPU and a Confidential TPU

    Google announced two new eighth-generation TPU system configurations: TPU 8i Max, which packages two TPU 8i chips into one package to double compute, SRAM and HBM bandwidth for long-context workloads, and TPU 8tx, which pairs TPU 8t silicon with 5th Gen AMD EPYC hosts for confidential inference with up to 2x lower latency than TPU 8t. Google published no absolute specs, benchmarks, prices or regions for either system, and both are reported to arrive in 2027. The announcement also covered software updates including open-sourced TorchTPU, vLLM and SGLang TPU backends, a TPU agent suite preview, and TPU Sync, which cut median KV transfer latency on Qwen3 235B from several seconds to about 230 ms on Ironwood (TPU v7x).

  19. Hacker News AI58

    Sleep wearables are training AI on your brainwaves without explicit consent

    An investigation by Science with the Pulitzer Center's AI Accountability Network found that neurology companies have accumulated tens of millions of hours of EEG recordings from sleep wearables and some are using them to develop brain foundation models, despite initially failing to spell out that use in their privacy policies. Interaxon, maker of the Muse sleep and meditation headband, has trained an AI model on about 50,000 hours of EEG recordings and says it removes identifying information such as names, emails and addresses before use. Its Chief Marketing Officer Nadia Kumentas said the company aims "to use as much of that eligible corpus as possible."

  20. Latent Space58

    What Frontier Labs Buy: A Data Deep Dive with Handshake AI — Garrett Lord & Team

    Latent Space's first deep dive with an AI data company examines Handshake's AI training business, which went from launch to $1 billion in gross annualized revenue in roughly a year. The episode walks through live demos of reinforcement learning environments across finance (Project Falcon's Excel grader), life sciences (the VIALS benchmark for scientific images), and safety/alignment simulations, with segments on how frontier labs buy and evaluate data.

  21. Hacker News AI43

    PwC plans to certify workers in critical thinking as AI changes their jobs

    PwC US is launching a certification program starting in December for more than 85,000 employees across its US and Mexico offices and Acceleration Centers to build and demonstrate critical-thinking and judgment skills, with completion targeted by the end of its fiscal year in June 2027. The roughly six-month program begins with an assessment of thinking style, followed by classes developed with professors and veteran staff, realistic simulations, and a personalized AI coach, and proficiency will be evaluated in year-end performance reviews rather than through a pass/fail exam. PwC said AI is taking over foundational tasks that once helped employees develop judgment, and a recent analysis of US job listings found entry-level roles exposed to AI were far more likely to include senior-level skills such as judgment and leadership than in 2019.

  22. The Decoder58

    Disney bets on Adobe's Firefly to create brand-safe AI models for its entertainment empire

    Disney and Adobe announced a multi-year partnership at Adobe's MAX conference, under which Disney gets early access to new Adobe technology for its theme parks, cruise line, and ESPN content. Adobe will use Firefly Foundry to build custom models trained on Disney characters, environments, and visual styles to produce commercially safe, production-ready output, while Disney's Imagineering division already uses Firefly for theme park development. The deal also covers Walt Disney Studios' StudioLAB and ESPN Edge, where AI is meant to boost fan engagement for sports content and unlock new uses for ESPN's archive.

  23. Latent Space58

    What Frontier Labs Buy: A Data Deep Dive with Handshake AI — Garrett Lord & Team

    Latent Space's first deep dive with an AI data company examines Handshake's AI training business, which went from launch to $1 billion in gross annualized revenue in roughly a year. The episode walks through live demos of reinforcement learning environments across finance (Project Falcon's Excel grader), life sciences (the VIALS benchmark for scientific images), and safety/alignment simulations, with segments on how frontier labs buy and evaluate data.

  24. Hacker News AI88

    AI R&D Is Accelerating Towards Potential RSI in 2027

    METR's deep-access investigation of Anthropic found that since October 2025 Anthropic's models have been improving on its internal AECI benchmark at roughly 1.9x the pre-October 2025 rate (95% CI 1.5–2.8x), with a majority of the acceleration likely attributable to AI doing R&D and no observed bottlenecks preventing further acceleration. METR reports that Anthropic redacted most of the evidence it found most informative, and critics such as Toby Ord argue the data are also consistent with a one-time level shift from the larger Mythos pretrain (implying ~1.4x). The post's author calls the situation grim and argues timelines may be shorter than the AI 2027 scenario, while noting the estimates could be wrong or bottlenecks could be hit.

    Why it matters: Summarizes a widely discussed METR report on AI-driven acceleration of AI R&D and the debate over whether it implies near-term recursive self-improvement.

Nov 9Mon
  1. Databricks Blog36

    Accelerate GPU training on Unity Catalog tables with UCDataset

    Databricks released UCDataset in Beta on AI Runtime, a native PyTorch data loader that streams Unity Catalog tables directly to every GPU in a training job without exports, copies, or custom loader code. On a simulated production workload streaming a 300 GB feature table across 8 NVIDIA H100 GPUs to train a recommendation model, UCDataset cut training time by 10x on average compared with loading data directly from Parquet. The Beta reads Delta tables in Unity Catalog, applies filters and deletion vectors as data is read, and enforces Unity Catalog permissions, though it cannot read tables with row filters or column masks.

  2. Tomasz Tunguz38

    When Does the US Become a Net Importer of Tokens?

    With AI demand outpacing supply and domestic data center construction slowing, the US is projected to become a net importer of tokens between 2028 and 2030, depending on how many planned facilities are cancelled. Current systems plus new data centers can handle roughly 2.5x annual token growth, and imports begin by 2028 if 45% of planned projects fail to connect, by 2029 at 20% cancellations, or by 2030 even if every proposed facility is built on time. The piece notes that demand has been price-inelastic so far, with short-term GPU rental prices rising from $5 to more than $13 per hour over two quarters while buyers kept paying.

  3. Hugging Face Blog36

    Nemotron-Personas-Finance-Japan: A Synthetic Dataset Capturing Japan's Demographics and Financial Context

    Elith and NVIDIA released Nemotron-Personas-Finance-Japan, an open synthetic persona dataset built on official Japanese statistics and household finance and financial literacy data, generated with NeMo SDG-PGMs and NeMo Data Designer. It contains 9M personas (1M records x 9 personas) in natural Japanese across 50 fields, released under CC BY 4.0 for commercial and non-commercial use. The dataset is designed for financial AI development such as conversational agents, customer simulation, model evaluation, and bias testing without using real customer data.

  4. The Verge · AI42

    Jonathan Kanter is using AI to take on Big Tech

    Former Justice Department Antitrust Division head Jonathan Kanter and former acting division chief Doha Mekki have launched FluxCap, a startup staffed with former DOJ staffers and Palantir engineers that sells an AI-based legal research and analysis tool to help plaintiffs build antitrust cases. The platform, built on proprietary technology and existing large language models, is trained to identify the facts and legal framework determining whether a case succeeds, find holes in attorneys' logic, and sift through discovery materials for signs of anticompetitive conduct. Kanter describes it as a "force multiplier" for enforcers, saying a small team gains the capacity of roughly "20-plus people working around the clock."

  5. MIT Technology Review · AI12

    Innovating with AI in the special chemicals industry

    Lubrizol VP of IT Girish Rao describes how the specialty chemicals manufacturer is building a digital foundation linking business processes, modern data platforms and cloud infrastructure to help R&D teams test ideas faster and develop more sustainable materials. Rao says data scientists can identify patterns and model possibilities before physical experiments, but the aim is "not about replacing scientific judgment," and the company uses "fusion teams" of scientists, data scientists and IT staff that start with small pilots to build trust. He expects AI, automation, advanced modeling, digital twins and connected data to converge as laboratories and plants become more intelligent.

  6. Cloudflare Blog30

    Launching the Vite Ecosystem Fund

    Cloudflare is distributing the first $301,300 of the Vite Ecosystem Fund to 44 projects and maintainers, part of a $1 million commitment made when VoidZero joined Cloudflare. The post argues that AI has made open source funding urgent by increasing AI-authored pull requests and maintainer burnout, and it describes the fund's selection panel and stipend structure of up to $4,000 per month at $50 per hour for eligible core maintainers.

  7. Exponential View22

    📈 AI-exposed workers wait a month longer for a job

    Anthropic paid $3.4 billion to take over a supplier's two-year-old order for 1.7GW of gas generators, the largest known power-equipment purchase by an AI lab. Apple's memory inventory grew 2.5x year on year, helping protect against memory cost increases. Workers who lost jobs in AI-exposed roles in 2025 took about a month longer than peers in other roles to find new work.

  8. Hacker News AI40

    Human feedback guides AI approach that automates about 84% of decisions

    Researchers Kai Hoberg, Christina Imdahl and William Schmidt propose "targeted automation," a four-step approach that automates decisions where human intervention adds little value while keeping people involved where their input can improve the algorithm. In an empirical application using more than 475,000 order-related decisions from a four-year period at a major commercial vehicle manufacturer, they were able to automate around 84% of decisions. The work was published in Production and Operations Management.

  9. Lovable Blog38

    How we moved our frontend off Firestore and built Lovable Realtime

    Lovable built its own WebSocket-based realtime service, Lovable Realtime, to replace browser-side Firestore listeners, letting its Go backend control payload shape and re-evaluate authorization continuously across Firestore and Spanner data. The service handles peaks of over 2 million concurrent subscriptions and delivered about 2.61 billion data and patch frames in a month, with update latency of 66 ms at p50 and 249 ms at p99. The rollout used shadow traffic and a gradual feature-flag ramp to 100% of users without a user-facing incident.

  10. NVIDIA Newsroom25

    NVIDIA Announces New Planned Investments in Korean Startups at the MSIT-NVIDIA AI Startup Partnership Forum

    Korea's Ministry of Science and ICT and NVIDIA hosted the MSIT-NVIDIA AI Startup Partnership Forum at COEX, where Vice Minister Ryu Je-myung and NVIDIA leaders met with over 20 Korean startups spanning foundation models, physical AI, AI infrastructure and AI semiconductors. NVIDIA announced an investment in Point2 Technology, an AI interconnect company building RF-based interconnects for AI cluster scale-up, with additional investments planned in Motif Technologies, RLWRLD and Upstage. The planned investments build on NVIDIA's work with Korea's startup community, including more than 1,000 Korea-based members of NVIDIA Inception.

  11. Hacker News AI42

    370 Books in 20 months: South African author defends her use of AI

    South African writer Coral Hart, a pseudonym, told AFP she has published more than 370 books since February 2025 using AI models including ChatGPT, Claude and Grok, and says she has received tens of thousands of hate messages and several death threats. She defines story concepts, characters and chapter outlines before her system drafts books, which then go through AI edits and a human proofreader, and she estimates she has sold over one million books and earned more than $1 million this year. Hart, who heads an operation training several hundred people in her method and is preparing software for AI authors called Authorythm, responds to "AI slop" accusations by saying "Books are selling."

  12. MIT News AI58

    A single AI model simulates the behavior of many substances, from sliding sand to folding proteins

    Researchers at MIT, UBC and elsewhere have developed WorldParticle, a deep-learning system that represents materials as Lagrangian particles and predicts their motion one step at a time, learning underlying dynamics rather than relying on hand-coded solvers. The model was demonstrated on six physical phenomena, including deformable objects, granular materials and proteins, with higher accuracy than numerical methods and modeled protein structural changes about 100 times faster than numerical simulators. The work will be presented at ACM SIGGRAPH Asia.

  13. Hacker News AI58

    Trump Official Wants Data Centers on Public Lands, Has $10M+ Invested in AI

    Interior Secretary Doug Burgum, the administration's point person on AI infrastructure, holds between $18.1 million and $64.4 million in AI company investments through Arthur Ventures, according to his 2026 financial disclosure analyzed by Save Our Parks for ProPublica. Burgum has promoted data centers on federal lands and met more than 20 times with Amazon, Microsoft and Alphabet representatives from January through November 2025, while ethics experts said he should recuse from such decisions. Other officials, including BLM Director Steve Pearce and USDA Undersecretary Michael Boren, also hold AI-related investments as their agencies review data center permits.

  14. Hacker News AI71

    Revolut Cut AI Costs 90% and Started Shipping Faster

    At AI Engineer New York, Revolut's AI operations lead David Abitbol described moving about 2,000 engineers off metered Anthropic and OpenAI APIs to Z.ai's open-weight GLM model, self-hosted on 56 Nvidia B200 GPUs via Together AI, at a blended $0.30 per million tokens and one-tenth the previous invoice. After the switch, lead time from picking up a feature to merge-ready code fell by about 70%. The piece also covers fleet tuning that absorbed 76% more demand on the same GPUs, compliance for a Chinese-developed model at an EU-regulated bank, and where value is moving in the AI stack.

  15. Artificial Analysis Articles42

    Introducing the Artificial Analysis Decision Index v2

    Artificial Analysis released Decision Index v2, with 45% of the Index built on private datasets (up from 25% in v1) and a new Decision Score that rewards both correctness and calibrated confidence. Drex Max v1.1 (Nace AI) and GLiDE v1.2 (Fastino Labs) both score 73, with Mammoth 27B (Baseten) at 72, while cost per 1M Tasks ranges from $14 for Celeris-1.1 Decision to $859 for Clef. The update adds AA-Paperclip (1,500 private documents), AA-Symmetry, revised AA-Omniscience-Decision and HLE-MultiChoice benchmarks, and EvalSafe-O*NET-AA, while reducing When2Call to 10% and removing WildGuardTest.

  16. Artificial Analysis Articles40

    Endpoint Accuracy Index v1.1: Tighter long context intervals and expanded model coverage

    Artificial Analysis released Endpoint Accuracy Index v1.1, which updates AA-LCR-25 to dataset version 1.1 with corrected answer keys, an updated equality-checker LLM and 25 repeats per question, narrowing the average 95% confidence interval from ±4.7 to ±2.6 points. Coverage now spans 103 serverless endpoints across six open weights models, adding GLM-5.3 Flash and Muse Spark Open and replacing DeepSeek V4 Pro with DeepSeek V4.1 Flash 1017, while GLM-5.2 leaves. Of the 103 endpoints, 24 score statistically below their self-hosted reference, with causes grouped as output token caps, reduced reasoning, tool call handling differences, and request limits or stream behavior.