OrangeBot.AI Digest — 2026-07-20

AI Coworkers that run your projects and operations

Keep your desktop clean by auto-organizing screenshots.

AI Crypto Investigations/Research Agents

Agentic click tracking that shows what grows your business

Understand anything online without leaving the page.

The live AI employee inside your product, serving every user

Open weights 975B multimodal model built for fine-tuning

We warm your prospects before reaching out

Turn drum audio into a score you can actually edit

Create AI ads that convert, not just look good

See how a brain reacts to your clip

The AI-native Operating System for founders.

Replay QA tells you what is broken before your users do

Your personal context file for every agent.

Build AI agents your whole team can run, and control

Skills are a useful abstraction for software agents, turning human and agent experience into reusable procedural knowledge. Yet existing skill libraries are mostly hand-written, text-centric, or derived from agent traces, leaving tutorial videos and other multimodal human resources largely underused. We present RESOURCE2SKILL, a framework that distills multimodal resources, including tutorial videos, repositories, articles, and reference artifacts, into executable skills for software agents. RESOURCE2SKILL organizes these skills as a hierarchical multimodal Skill Wiki, where each entry combines structured text, code, visual examples, metadata, and provenance. This design preserves complementary signals from different resources: videos capture temporal operations and visual effects, code captures executable tool patterns, and articles or artifacts provide conceptual and stylistic grounding. At inference time, agents retrieve and compose relevant skills from the wiki; when coverage is insufficient, the same construction operator can acquire new skills online. Across seven practical authoring domains, RESOURCE2SKILL improves average overall score by +11.9 percentage points over no-skill agents and outperforms strong harness baselines in 26 of 28 main-aggregate model-domain cells. Ablations confirm the value of multimodal skill format, hierarchical organization, source diversity, selection strategy, and online acquisition.

Graph retrieval-augmented generation (GraphRAG) enhances large language models with structured knowledge, yet existing systems construct knowledge graphs in a single extraction pass, producing noisy entities and brittle retrieval. RAGU, an open-source modular GraphRAG engine, addresses this by separating extraction from consolidation: entities and relations pass through two-stage typed extraction, DBSCAN-backed deduplication, LLM summarization, and Leiden community detection. A key insight motivates a compact extractor: the skills an in-pipeline LLM needs - comprehension, extraction, reasoning over context - are language skills that grow only weakly with model size, unlike factual world knowledge. Accordingly, we train Meno-Lite-0.1, a 7B model optimized for language skills, which outperforms Qwen2.5-32B on knowledge-graph construction (+12.5% relative harmonic mean) and matches it on English GraphRAG tasks. On GraphRAG-Bench (Medical), RAGU retrieves the most complete context at every factoid level (evidence recall up to 0.84 vs. leq0.76) and overtakes HippoRAG2 on synthesis tasks; on multi-hop factoid QA, the apparent HippoRAG2 advantage is shown to be largely an answer-format artifact. RAGU is installable via pip install graph_ragu, runs on a single GPU, and is released under MIT. The source code is publicly available at https://github.com/RaguTeam/RAGU, and the Meno-Lite-0.1 model can be obtained from https://huggingface.co/bond005/meno-lite-0.1.

We present Xiaomi-Robotics-1, a foundational vision-language-action (VLA) model capable of (1) following diverse language instructions to perform a wide range of mobile manipulation tasks in unseen environments out-of-the-box, and (2) efficiently adapting to novel downstream tasks with minimal fine-tuning data. We propose a two-stage training recipe consisting of pre-training and post-training. During pre-training, we imbue the model with broad and generalizable action-generation capabilities by training on over 100k hours of real-world manipulation trajectories collected via UMI devices. Crucially, we develop a scalable auto-labeling pipeline that annotates trajectory clips with natural languages describing scene state transitions, providing rich and precise conditioning for action learning. During post-training, we aim to align these capabilities with robot embodiments and imperative instructions that humans naturally use to prompt robots. Extensive experiments demonstrate strong scaling behavior. Xiaomi-Robotics-1 consistently improves with increased data scales and model sizes during pre-training. This scaling behavior directly transfers to post-training, where a stronger pre-training model yields better out-of-the-box real-robot performance in unseen environments. Furthermore, Xiaomi-Robotics-1 serves as a strong robot foundation policy that can be efficiently fine-tuned on complex, dexterous tasks with high data efficiency. Across multiple simulation benchmarks, Xiaomi-Robotics-1 outperforms state-of-the-art methods. Notably, it establishes a new state-of-the-art with a 57.6% success rate on RoboCasa365, surpassing the previous best of 46.6%. Furthermore, it achieves an average score of 20.07 on RoboDojo, significantly outperforming the prior state-of-the-art (13.07). Code and model checkpoints will be released. Project page: https://robotics.xiaomi.com/xiaomi-robotics-1.html

We present Loopie, the most powerful looped Transformer to date. The Loopie series consists of two Mixture-of-Experts (MoE) models: a 20B-parameter model with 2B active parameters and a 6Bparameter model with 0.6B active parameters. Looped Transformers have long faced a challenge: given an N-fold increase in pre-training compute, increasing the parameter count by a factor of N usually outperforms looping a model N times. Loopie addresses this challenge. Extensive ablation studies, including comparisons with a vanilla 30B-A3B model, show that Loopie substantially outperforms vanilla Transformer baselines trained with the same compute budget. Our novel post-training pipeline equips Loopie with strong reasoning abilities. At the 2025 IMO and IPhO, Loopie achieves gold-medal performance without tools.

Hyper-Connections (HC) expand the residual stream of Transformers into N parallel streams, providing a form of memory scaling beyond model width and depth. Manifold-Constrained HC (mHC) stabilizes this formulation at scale. The large gains from N{=}1 to N{=}4 suggest residual-stream expansion as a promising scaling axis. However, existing HC-family methods typically stop at N{=}4. Our experiments reveal why: scaling mHC beyond this point yields diminishing performance gains and rapidly increasing training cost. We attribute this limitation to two bottlenecks: insufficient write-back information for an expanding number of streams and residual-mixing generation whose cost scales cubically with N. To address both bottlenecks, we propose xHC (Expanded Hyper-Connections), the first HC-family method to achieve meaningful expansion beyond N{=}4. xHC combines temporal feature augmentation for richer write-back with a sparse residual-stream architecture that updates only k=4 of the N=16 streams while retaining dense access to the full residual state. Across 18B and 28B MoE models, xHC delivers strong and consistent downstream improvements. On an 18B MoE model, xHC improves the average downstream score by 4.0 points over mHC, while adding only modest training FLOPs over the vanilla baseline. Scaling-law experiments show that the vanilla and mHC require 1.50times and 1.19times the compute of xHC, respectively, to reach the same loss. Practical large-N training also requires controlling memory traffic from the expanded residual state. We therefore introduce xHC-Flash, which reduces the per-sublayer memory traffic from 73.5C to 40C, comparable to the 34C required by mHC at N{=}4, while retaining the gains of full xHC. Together, xHC and xHC-Flash make large-N residual-stream expansion effective and practical for LLM pre-training.

Healthcare spans high-stakes communication, expert reasoning, and workflow execution, yet specialized LLMs that cover these use cases together remain limited. A healthcare model must handle patient consultation, clinical reasoning over text and images, interactive diagnosis, and electronic health record (EHR) tool use. These capabilities fail in different ways, and a narrow update for one task can degrade another. We present Cura 1T, a healthcare-specialized LLM trained through a human-gated self-evolution loop. In each evolution round, a training agent plans a target capability, trains the model, evaluates benchmark trajectories, and refines the data mixture from observed failures. This data-centered loop improves the model through targeted synthetic and curated examples rather than a single generic medical-data update. Across the healthcare evaluation suite, Cura 1T ranks at or near the top among frontier baselines, while remaining competitive on out-of-domain reasoning and agentic benchmarks.

On-policy distillation is an alternative post-training method in reinforcement learning that alleviates the constraints imposed by reward models by providing token-level supervision from a teacher model. Although on-policy distillation has been studied and applied across various settings, its fundamental design remains underexplored. In this paper, we introduce a new distillation reward, termed the delta signal, instead of directly imitating the teacher's output distribution. The delta signal is defined as the difference between the teacher model and its base model prior to instruction tuning for reasoning capability. It therefore captures the changes induced by reasoning tuning and provides a more direct signal for transferring reasoning capabilities. Using extensive empirical evidence, we show that the delta signal substantially improves on-policy distillation and refer to the new distillation method as On-Policy Delta Distillation (OPD^2). Experiments across mathematics, science, and code-reasoning benchmarks demonstrate that OPD^2 consistently outperforms conventional on-policy distillation, enabling reasoning LLMs to achieve strong performance with only a short post-training period. Code will be available at https://github.com/naver-ai/opd2

Large language models (LLMs) are transforming recommender systems from matching co-occurrence patterns in historical behavior toward reasoning about the intent that drives it. RecGPT-V1 pioneered this paradigm on Taobao by centering user understanding, and RecGPT-V2 scaled it via coordinated multi-agent reasoning; both are deployed in production with consistent gains in user experience and commercial outcomes. However, operating RecGPT at scale reveals three challenges: (1) stateless behavior modeling, where each request reprocesses full user history, wasting computation and discarding prior analysis; (2) a tag-to-item information bottleneck, where natural-language tags form a lossy channel between user understanding and item grounding; and (3) inefficient explicit reasoning, whose lengthy chain-of-thought incurs untenable latency and compute overhead. We present RecGPT-V3, a stateful, hybrid-modal recommender that reasons over natural language for open-world knowledge and Semantic IDs (SIDs) for concrete item grounding. A Memory Hub maintains structured, continually evolving user memory that distills long-horizon behavior into condensed units, cutting user-modeling computation by 55.8%. A Hybrid-modal Foundation Model allows the LLM jointly reason over text tags and SIDs, opening a high-bandwidth channel into the item space. Latent Intent Reasoning internalizes verbose rationales into compact learnable latent tokens that remain decodable into readable explanations, lowering output token cost by 200x. Deployed in Taobao's "Guess What You Like" feed, RecGPT-V3 achieves consistent gains in large-scale online A/B tests: IPV +1.28%, CTR +1.00%, TC +1.97%, GMV +3.97%, while cutting end-to-end serving resource consumption by 52.4%.

Code review helps maintain software quality before code integration, but it also imposes a substantial workload on human reviewers. As generative artificial intelligence becomes part of software development, code review is shifting from a primarily human review process toward AI-supported review processes in which large language model (LLM) reviewers and AI agent reviewers participate alongside human reviewers. However, we still lack empirical evidence on how this transition affects review efficiency and review quality. In this paper, we study 1.02 million reviewed pull requests from 207 GitHub projects that transition across three code review eras: human-centric review, LLM-assisted review, and agentic code review. We identify three AI reviewer adoption practices: Gradual AI Adoption, Rapid LLM Adoption, and Rapid AI Agent Adoption. We further model pull request review discussions as reviewer interaction sequences to characterize how human, LLM, and AI agent reviewers collaborate during the review process. Our results show that agent-involved collaboration patterns, especially reviews initiated by AI agents or involving multiple AI agents, are associated with faster review decisions under Gradual AI Adoption and Rapid AI Agent Adoption. However, these efficiency gains do not translate into better review quality. We also find that review activity and pull request type remain important across eras, while human-AI collaboration patterns become the strongest explanatory factor for review efficiency once LLM and AI agent reviewers participate. These findings provide empirical guidance for designing AI-supported code review processes that improve efficiency without weakening review quality.

In this report, we introduce Qwen-Music, a powerful music generation model capable of producing highly musical and high-fidelity songs with complete vocal singing. Qwen-Music supports two core tasks: Text to Music Generation, which create entirely new songs from text descriptions, lyrics, and musical attributes, and Cover Song Generation, which reinterprets existing songs with different styles and vocal characteristics. Architecturally, Qwen-Music integrates three core components: Qwen-Music-Tokenizer, Qwen-Music-LLM, and Qwen-Music-Render. Qwen-Music-Tokenizer compresses audio into a 25 Hz single-codebook stream of Music Semantic Tokens that preserve semantic and melodic information for LLM prediction. Based on these tokens, Qwen-Music-LLM performs autoregressive music semantic modeling, with a key novelty being a melody-token-based chain-of-thought (Melody-CoT) mechanism that plans melodies before full-song generation, improving creativity, musicality, structural coherence, and reference-audio-based melody cloning. To overcome the fidelity limitations of discrete semantic tokens, Qwen-Music-Render performs generative stereo rendering, enriching acoustic details and producing high-fidelity stereo waveforms. Finally, we train Qwen-Music-LLM on more than 5 million hours of multilingual music data covering hundreds of languages. We first apply quality-aware pre-training curriculum, then use progressive post-training, comprising supervised initialization, offline DPO, and online GSPO, to further improve musicality and instruction-following ability. Across 600 Chinese and English prompts, Qwen-Music achieves state-of-the-art results in 13 of 16 objective musicality and audio-quality metrics. Professional evaluators also prefer Qwen-Music over leading proprietary systems. For cover song generation, Qwen-Music preserves reference melodies more accurately than leading proprietary systems.

Reinforcement learning (RL) has become central to improving large language models (LLMs) on complex reasoning tasks, yet RL post-training is largely studied in isolation from the pretraining that precedes it. As a result, two basic questions remain open: (1) how do pretraining choices (model size, data) shape the returns to RL compute, and (2) what does RL actually do to the model? These questions are difficult to study in the standard LLM setting: pretraining corpora are vast and uncontrolled, making it hard to attribute behaviors to pretraining versus RL, and systematic compute sweeps across both stages are prohibitively expensive. To address these challenges, we use chess as a controlled testbed for studying reasoning across the full pretraining-to-post-training pipeline. We follow the standard LLM training pipeline by pretraining language models from 5M to 1B parameters on human chess games, supervised fine-tuning on synthetic reasoning traces, and running RL on chess puzzles with verifiable rewards. Using this framework, we find that the post-RL performance at given RL compute level is well-predicted from the pretraining loss, and slope of the RL reward curves improves approximately linearly with the pretraining tokens. Beyond scaling, we find that RL does not simply sharpen the SFT policy: on easy puzzles it amplifies correct moves the SFT policy already preferred, while on hard puzzles it surfaces correct moves that were nearly absent under SFT. We further test whether our findings transfer beyond chess by training a 1B language model on math-domain text, where the same predictive pattern emerges: longer-pretrained checkpoints reach higher post-RL performance and improve faster under RL. In sum, we provide a quantitative account of the pretraining-to-RL interface and a controlled testbed for studying the science of reasoning across the full pretraining-to-post-training pipeline.

Under model--harness co-evolution, harnesses are not merely inference-time scaffolds but data-generating components whose execution traces can shape future foundation models. This motivates harness-in-the-loop learning: optimizing harnesses for both immediate agent performance and the quality of traces used for future model training. However, continually updating provider-built scaffolds is costly and labor-intensive. We therefore investigate whether optimizing user-constructed harnesses in a task-specific manner can improve execution-trace quality while remaining computationally lightweight and requiring only a few update iterations. To this end, we introduce Recursive Harness Self-Improvement (RHI), which represents the harness as a prompt-level specification of the agent loop and iteratively refines it using pairwise feedback over its own revision history. Across 30 synthetic machine-learning research tasks spanning quantitative finance, robotics, and pharmacy, a few RHI iterations suffice to substantially raise the performance ceiling of low-reasoning-effort agents, exceeding the corresponding maximum-reasoning-effort setting while reducing inference cost by up to 60%. We show that these gains arise primarily from improved task-specific context management through more effective inter-agent information flow rather than longer reasoning traces. Finally, we formalize this behavior as an information-theoretic hypothesis for RHI's implicit optimization objective, suggesting RHI as a practical algorithm for continual learning within the paradigm of model--harness co-evolution.

Muon is competitive with AdamW in large-scale pre-training, but its value for reinforcement-learning (RL) post-training remains unclear. We study vanilla Muon in sparse-reward agentic RL through matched single-seed comparisons with AdamW on ALFWorld using Qwen2.5-0.5B-Instruct. Under Group-in-Group Policy Optimization (GiGPO), applying Muon only to hidden weight matrices raises final-window validation success from 0.290 to 0.546 (+88%); high-rate AdamW controls retain no post-update success. The effect depends on the advantage estimator and learning rate. At 3e-5, Muon improves GRPO from 0.161 to 0.268, whereas GraphGPO's late-window gap narrows near saturation. At 1e-5, GraphGPO Muon reaches 0.901, raises normalized validation AUC from 0.399 to 0.556, and reaches 0.5 and 0.75 success 30 and 60 updates earlier, respectively. These exploratory results show that Muon can benefit agentic RL and motivate studying the policy optimizer, advantage estimator, and learning rate jointly. Multi-seed and cross-task validation remain open.

Reinforcement learning with verifiable rewards (RLVR) commonly uses entropy for advantage shaping. However, entropy cannot distinguish useful uncertainty from detrimental confusion, limiting its effectiveness as a correctness signal. We propose Contrastive Policy Optimization (CPO), which uses token-level contrastive disagreement between reference-guided and vanilla generation distributions for correctness-aware advantage shaping. Both theoretical and empirical results show that this disagreement reliably indicates token-level correctness. We further show that On-policy Distillation is a special case of CPO, where the posterior distribution is instantiated by an external teacher model. CPO also resolves the zero-advantage problem. Experiments on in-domain and out-of-domain benchmarks demonstrate that CPO substantially outperforms entropy-based RLVR methods while maintaining strong generalization. Further analysis shows that correct and incorrect responses naturally support exploration and exploitation respectively, and balancing both leads to the best performance.

We present S1-Omni, a unified multimodal reasoning model for scientific understanding, prediction, and generation. AI for Science (AI4S) has advanced significantly through domain-specific models, tool-augmented LLMs, and scientific language models. However, model capabilities remain highly fragmented, limiting the joint modeling of heterogeneous data, scientific laws, and expert knowledge. S1-Omni addresses this gap by consolidating these capabilities into a single, coherent scientific reasoning model. The architecture of S1-Omni is built upon three core components: unified representation of scientific data, natural-world knowledge alignment, and decoding for domain-specific tasks. First, S1-Omni maps natural-language instructions and scientific objects, including CIF, SMILES, protein sequences, spectra, and scientific images, into a shared representation space. Second, it incorporates scientific laws and expert knowledge into data construction and training, enabling the model to reason from scientific evidence. Third, it performs task-specific decoding to support a broad range of applications, including property prediction, spectrum-to-molecular generation, protein site and structure prediction, and scientific image generation and editing. S1-Omni is trained on S1-Omni-Corpus, which covers 200 scientific tasks and contains millions of reasoning samples, and is evaluated on over 60 scientific benchmarks. It outperforms GPT-5.5 and Gemini-3.1-Pro on most benchmarks and matches or surpasses domain-specific models on several benchmarks. Overall, S1-Omni provides a practical path toward unified scientific modeling.

Blake Brittain / Reuters : Judge approves Anthropic's $1.5B settlement with authors, the first major US case of its kind to settle; some authors opted out and have other ongoing lawsuits —  A federal judge in San Francisco on Monday signed off on artificial intelligence company Anthropic's landmark $1.5 billion settlement …

Bloomberg : Source: Ancestry.com pitches a risky $1.75B term loan at 4 to 4.25 percentage points above the benchmark to refinance debt from its 2020 Blackstone acquisition —  Ancestry.com Inc. is seeking to refinance debt raised to fund its 2020 acquisition by Blackstone Group Inc. funds …

Bloomberg : Snap reaches a deal to settle the second case set for trial over social media harm to minors, following TikTok and YouTube, leaving Meta as the only defendant —  Snap Inc. said it has reached a tentative agreement to settle the second case set for trial over claims that the world's largest …

Sarah Perez / TechCrunch : X relaunches its Android app after a year-long rewrite from scratch; Nikita Bier calls it “one of the largest engineering projects” in the company's history —  Nearly a year ago, Elon Musk-owned X announced it would begin rebuilding the Android version of its app …

Cheng Ting-Fang / Nikkei Asia : Taiwan indicts a former TSMC executive for allegedly stealing trade secrets, in what authorities say is the first National Security Act case involving China —  TAIPEI — Taiwanese prosecutors on Monday indicted a former TSMC deputy manager for allegedly stealing trade secrets involving …

Marina Temkin / TechCrunch : Natural, which aims to help companies let their AI agents make autonomous payments, raised a $30M Series A led by Forerunner, bringing its total raised to $40M —  AI agents are starting to execute more sophisticated tasks, such as identifying vendors that can deliver freight, comparing prices …

OpenAI : OpenAI paused internal access to an unreleased model that disproved the Erd&odblacs unit distance conjecture after it repeatedly found ways to act outside its sandbox —  What internal use of a long-running model taught us about safety.  —  Summary  — Long-running models can solve difficult …

Reuters : Anduril and Archer unveil an autonomous aircraft platform designed for commercial and military uses, along with an autonomous attack rotorcraft called Thunder —  Archer Aviation (ACHR.N) and defense technology company Anduril unveiled a co-developed autonomous aircraft platform on Monday …

Lorenzo Franceschi-Bicchierai / TechCrunch : Cybersecurity companies say hackers are exploiting vulnerable WordPress versions to take over websites; WordPress patched two critical security flaws last week —  Hackers are breaking into websites that run vulnerable versions of the popular blogging software WordPress, according to several cybersecurity firms.

Bloomberg : Source: Z.ai completed construction of a 1 GW data center housing only Chinese chips; Z.ai has built or operates several computing clusters each with 10K+ chips —  Z.AI has completed construction of a giant data center that houses only Chinese-made chips, a big step in Beijing's efforts …

Chris Metinko / Axios : Chicago-based Empirical Security, which uses AI to help companies predict threats by monitoring exploited vulnerabilities, raised a $25M Series A —  Exposure management startup Empirical Security raised $25 million in Series A funding led by Brightmind Partners, CEO Ed Bellis tells Axios Pro.

Matthew Sparkes / New Scientist : Levent Alpöge, a mathematician who works at Anthropic, says he was able to disprove the 87-year-old Jacobian conjecture with the help of Fable 5 —  Levent Alpöge announced the solution on X A mathematician has cracked an 87-year-old conundrum with the help of AI and announced the solution unceremoniously in a tweet.

Dominic-Madori Davis / TechCrunch : Infinity, founded by Jeremy Nixon, the creator of hacker network community AGI House, to build an inference library that runs on all chips, raised a $15M seed —  AI infrastructure company Infinity announced a $15 million raise at a $100 million valuation on Monday from investors including Touring Capital …

Maria Curi / Axios : CAISI Director Chris Fall is resigning after taking over the federal AI testing institute in April; NIST Director Arvind Raman will serve as acting director —  Chris Fall, the director of the Center for AI Standards and Innovation, is resigning just three months after taking …

Golan Hazani / CTech : Source: Tencent is in talks to acquire Israeli game developer SuperPlay in a deal valuing SuperPlay at $1B-$1.5B; Playtika acquired SuperPlay in 2024 for $690M —  Exclusive: The transaction would transfer a fast-growing gaming studio, and a ballooning earnout commitment, to the Chinese technology giant.

医学界完成了一项前沿手术—— 一名男性通过移植自己童年取出并冷冻保存16年的睾丸组织，重新恢复了精子生成能力。这批睾丸组织样本是他在接受化疗前采集冷冻的。化疗会严重损伤生殖功能，导致不育。这一里程碑式成果有望开启全新的人类不育治疗浪潮。2008 年比利时布鲁塞尔自由大学医院的医生为一名患病的 10 岁男孩摘除部分睾丸组织并冷冻保存。他当时罹患镰状细胞贫血，即将接受化疗，为后续造血干细胞移植做准备。10 余年后，这名男孩成年，希望拥有自己的后代。医护人员对其进行了两年监测，确认他已完全无法正常生成精子。随后，团队将冷冻保存 16 年的11个睾丸组织片段，分别移植至其剩余睾丸内部及阴囊皮下。移植组织在成年男性的激素环境与生理条件下培育生长一年。一年后，研究团队取出移植组织并开展检测分析。检测结果显示，多个移植组织中均发现生精干细胞，且存在活跃的精子生成迹象。在其中一个移植样本中，研究人员成功检出一枚成熟精子。

Mozilla 释出了 Firefox 153.0。它是 ESR 版本，将提供至少 15 个月的支持。Firefox 153.0 主要新特性包括：Vulkan 视频解码支持，Linux 版支持 Video Acceleration API (VA-API) ，增加对 JPEG-XL 图像格式的支持，Windows 版支持 HDR 视频播放，PDF 功能增强，等等。

黑客入侵了罗马尼亚的土地管理机构 Agenția Națională de Cadastru și Publicitate Imobiliară（ANCPI)，在勒索失败之后删除了整个土地登记数据库以及备份，导致全国房地产市场瘫痪，官方应用和网站已离线一周。公证员无法记录新的交易，公民也无法获取所有权证明或详细的土地记录。黑客是在 7 月 14 日删库的，一天后 ANCPI 部分被盗数据在黑客论坛上出售，泄露的数据包括员工凭证、内部文件以及该机构的 IT 网络详细信息。ANCPI 表示正在重建系统。虽然黑客声称删除了备份，但 ANCPI 表示它还有备份副本。

体细胞突变随年龄增长而积累，能导致细胞死亡。根据发表在《npj Aging》期刊上的一项研究，科学家开发出一种方法量化体细胞突变对人类寿命的影响。研究只考虑体细胞突变，没有考虑其它任何衰老特征。研究发现，肝脏等增殖性组织通过细胞更新能在数千年内维持功能，而神经元和心肌细胞等有丝分裂后细胞则是长寿的重要瓶颈。综合多个器官预测的人类寿命中位数是 146-194 岁，约为目前人类寿命的两倍。这意味着体细胞突变是衰老的重要驱动因素，但无法单独解释目前的人类寿命，驱动衰老的还有其它因素。

法国和意大利研究人员发现，有了 AI 之后人们愈来愈不愿说不知道了，AI 抑制了人们的批判性思维，使得人们更容易重复 AI 生成的虚假信息。研究人员将参与者分成两组，一组必须在没有 AI 的帮助下回答问题，另一组可以向 AI 寻求建议。结果发现没有 AI 帮助时 44% 的人会回答说不知道答案，而有了 AI 的帮助后只有 3% 的人这样做。研究人员让参与者使用了一个更可能出错的模型 Step 3.5 Flash。结果显示，不使用 AI 正确率为 27%，使用 AI 后降至了 9%，原本会给出正确答案的人在听取了 AI 建议后给出了错误答案。尽管正确率下降了，但参与者的自信心却增加了。

Google 逾 4500 名员工联名写信给首席执行官皮查伊，要求公司给员工更明确的裁员保障。员工提出的要求包括：保证遣散费、在强制裁员前先推出自愿离职计划，以及取消据称按配额评定绩效的制度。请愿书由 Alphabet 员工工会发起，已送到 Google 总部。工会主席、软件工程师考尔（Parul Koul）说，Google 市值已达到约 4 万亿美元，公司在取得空前成功的同时仍不断裁员，说明管理层把利润放在员工之上。请愿书要求，Google 在所有产品部门裁员前，先让员工选择是否自愿离职，并保证被裁员工能拿到遣散待遇，或允许他们把遣散费换成更长时间的带薪休假。员工也要求停止按固定比例评出低绩效员工的做法，认为这种制度可能变成裁员工具。考尔说，员工到皮查伊办公室递交请愿书时，大门紧闭，管理层没有正式回应。他们最后把请愿书交给一名职员，对方承诺会转交给皮查伊。

国家航天局小行星监测预警研究中心首席科学家李明涛领导的团队在一篇论文中讨论了中国自己的行星防御系统。NASA 执行双小行星重定向测试(DART)任务的探测器于 2022 年 9 月撞击了小行星 Dimorphos，这是世界首次行星防御技术演示。中国的计划更雄心勃勃，旨在直接破坏小行星的结构，而不只是偏转其轨道。中国计划中的探测器预定在 2029 年或 2030 年以 26 倍音速以上或每秒逾 9 公里速度撞击目标——相比之下 DART 撞击速度为每秒 6.1 公里。此次任务包括四个目标：精确撞击小行星、成功偏转其轨道、精确测量撞击结果以及证明撞击并未对地球造成损害。目标小行星是 2015 年发现的 2015 XF261，直径约为 30 米。中国计划发射两艘探测器，一艘撞击，另一艘观察。

MPEG-4 是一套用于音频、视频信息的压缩编码标准，由 MPEG 制定，它由一系列子标准组成，其中 Part 2 定义了一个对各种视觉信息（包括自然视频、静止纹理、计算机合成图形等等）的编解码器，如 XviD 编码就属于 Part 2。7 月 19 日 MPEG-4 Part 2 部分的最后一个专利 BRPI0109962B1 过期，从此 MPEG-4 Part 2 专利全部过期。但 XviD 编码已经很少被今天的视频使用，而 AAC(MPEG-4 Part 3) 和 AVC/H.264(MPEG-4 Part 10)仍然存在有效专利。

当你在计算机上保存文档时，你实际上是在选择一种格式，一种计算机能理解的语言规则。这套规则决定了文字、表格、图像和格式指令的存储方式，以及它们在未来如何被检索、共享和读取。Microsoft Office 使用的格式几十年来一直是办公文档的主流格式，最早它使用二进制格式，扩展名为 DOC 和 XLS；Office 2007 后使用了基于 XML 的格式 DOCX、XLSX 和 PPTX。数亿人在使用这些格式。它们是世界各地办公室、学校、公共管理机构和法院的通用语言。微软的格式很多方面都是私有的，意味着它们属于微软，由微软控制，服务于微软的利益，而这些利益可能与用户的利益并不一致。开放格式——其规范公开、可免费获取且任何软件均可不受限制实现的格式——以一种任何兼容应用都能忠实读取、写入和重现信息的方式存储信息。私有格式则可能包含未公开的功能、私有扩展或只有原始软件才能正确实现的行为。其它应用能打开微软的文档，但并非总能忠实重现文档内容。这就是排他性的“锁定”。

资产管理服务商 Lansweeper 的数据显示，其监测的 Windows 设备仍然有 16.9%（六分之一）的设备运行 Windows 10。一年前 Windows 10 比例还高达五成，去年 10 月微软终止对 Windows 10 的主流支持后其比例开始快速下降，6 月降至 18.6%，但 Lansweeper 警告迁移速度正大幅放缓，中小企业最为显著。中小企业中仍有 21.4% 的设备运行 Windows 10 系统，成本是企业继续​​使用旧操作系统的主要原因。部分行业的受影响程度更高，如 23% 的医疗保健和制药系统仍在使用 Windows 10，而消费和零售设备中这一比例约为 22.7%。Windows 10 设备平均有 1903 个活跃 CVE 漏洞，Windows 11 设备只有 652 个，两者相差 2.9 倍。Lansweeper 指出，容易迁移到新版本操作系统的设备都已经完成迁移，剩下的都是最难迁移的设备。

根据发表在《Neurology》期刊上的一项研究，多种广泛使用的代糖与记忆力和思维能力随时间推移而加速下降相关。代糖摄入量越大，认知能力下降速度越快，这种关联糖尿病患者最为显著。研究针对了 12,772 名巴西成年人，平均年龄 52 岁，他们接受了八年的监测，填写了详细的食物调查问卷。研究人员根据其食物中的代糖摄入量分成三组，参与者还在研究期间接受了认知评估。结果显示，在考虑了年龄、性别、高血压、心血管疾病等因素之后，摄入代糖/甜味剂最多的人其整体思维和记忆能力的下降速度比摄入量最少的人快 62%，这相当于额外衰老了约 1.6 年；中等摄入量组的认知能力下降速度比最低摄入量组快 35%，相当于额外衰老约 1.3 年。阿斯巴甜、糖精、安赛蜜、赤藓糖醇、山梨糖醇和木糖醇六种代糖与与认知能力下降相关，塔格糖（Tagatose）是唯一与认知能力下降无关的甜味剂。

2024 年 3月 29 日，一位微软工程师在家执行例行测试时，发现登录测试机的时间比往常慢了约半秒。大多数人都不会认为这是什么大问题。但他却深入展开了调查，在流行压缩工具 XZ Utils 中发现了一个故意植入的隐蔽后门。有人花了两年时间才把后门植入到该工具中。Adrian Mastronardi 发表了一本关于 XZ 后门事件的书《Half a Second》，在非商业使用、禁止衍生的 CC 许可证（CC BY-NC-ND 4.0）下免费提供。本书讲述了 一位精疲力竭的志愿者独自维护着代码，被耐心而巧妙的操纵，最终交出了代码的维护权；一位工程师凭借一连串的运气和来之不易的直觉，在半秒钟的好奇心驱使下发现了这次攻击；以及构建该后门的幕后操纵者，此人至今身份不明，其身份可能永远也无法被揭露。

反数据中心抗议者在美国 42 州举行了 142 场抗议活动。这一抗议行动获得了两党民众支持、由前茶党领袖联合创办的组织 HumansFirst 协调。抗议者集会反对 HumansFirst 所称的数据中心“缺乏问责”的扩张以及“对自由不可接受的侵犯”。在弗吉尼亚州 Fredericksburg，一位年轻人举的牌子上面写着：“我们靠水而不是数据生存。”亚特兰大抗议者的标语是“佐治亚州从未投票支持超大规模人工智能数据中心”。HumansFirst 表示暂无全美和各大城市抗议者人数的统计数据。反对数据中心建设是少数几个能跨越意识形态团结美国人的议题之一。民调显示，只有三分之一的美国人认可数据中心的建设速度，只有 14% 的受访者支持在其社区建设数据中心。

法国持牌博彩/赌博游戏监管机构 Autorité Nationale Des Jeux 命令 ISP 屏蔽最大预测市场平台 Polymarket，任何被发现宣传该非法赌博网站的人都可能被处以最高 10 万欧元的罚款。ANJ 此前已经限制法国居民在 Polymarket 上进行任何金融交易，但它发现法国居民绕过了限制，Polymarket 的法国访客仍然持续增长。ANJ 希望加大对 Polymarket 的打击力度，再次强调该平台被认定为非法赌博网站。西班牙政府也下令封锁了预测市场平台 Polymarket 和 Kalshi。

2025 年美国新车仅有 0.6% 是手动挡，汽车已是自动挡的天下，电动汽车和自动驾驶汽车的兴起也加速了手动挡汽车的消失。相比下欧洲的情况略好，行业分析公司 JATO Dynamics 的数据显示，欧洲最大汽车市场手动挡汽车的注册比例从 2001 年的 91% 下降到 2024 年的 29%。曾以手动挡汽车知名的斯巴鲁（Subaru）已停止提供手动挡车型，大众汽车将于今年停产在美国销售的最后一款手动挡车型捷达 GLI。丰田、本田和宝马都减少了面向美国市场的手动挡车型。