OrangeBot.AI Digest — 2026-08-03

Search every AI Agent & IDE Conversation

Create interactive 3D and XR experiences without coding

Cloud canvas where your team and AI agents works together

Animated SVG mascot studios for apps that need a personality

Make existing Codex and Claude Code sessions multiplayer

Wild Pokémon appear while you wait for Claude Code

Any harness, any LLM — cloud-hosted agents on demand.

Build campaigns, optimize spend, and create reports.

Qwen’s most capable model for coding and cowork

Own your voice again. Local-first voice & meeting recorder.

AI Faceless Video Generator for TikTok & YouTube

Turn sign language into speech with smart glasses

Play live electronic music with your hands using your webcam

Clone any Mac app into a fully isolated instance

Your subconscious mind as a living 3D garden you water daily

Reinforcement Learning with Verifiable Rewards (RLVR) has driven recent progress in reasoning-oriented large language models (LLMs) by enabling large-scale optimization. However, its applicability remains largely limited to domains such as mathematics and coding, where correctness can be deterministically verified. Open-ended tasks instead often rely on human preferences, reward models, or LLM-based judges, introducing evaluation bias, judge capability bottlenecks, and additional inference costs.Drawing on the principle of self-supervised learning, which constructs pretext tasks to derive supervision from the data itself, we propose Reinforcement Learning with Self-Verifiable Rewards (RLSVR), a task-transformation-based training paradigm for extending RLVR to open-ended tasks. RLSVR transforms open-ended tasks into verifiable proxy environments whose internal rules and interaction outcomes automatically generate reward signals. We instantiate RLSVR with SpyRL, a multi-agent self-play environment inspired by Who Is the Spy?. Agents receive asymmetric information, complete the same target task, and vote to identify a designated spy. Because the spy identity is predetermined, voting outcomes provide fully verifiable rewards, while successful identification remains closely related to output quality. Experiments on text summarization, creative writing, and mathematical reasoning show that SpyRL outperforms existing self-improvement methods on non-verifiable tasks and yields consistent gains on verifiable reasoning tasks. These results demonstrate that task transformation can extend scalable RLVR-based self-improvement beyond inherently verifiable domains. Models and code have been released at https://github.com/wangqinsi1/SpyRL.

World models enable a predictive substrate for planning and action, yet existing formulations merely answer a physical question: what/where it is, and how will it evolve. Human behavior, however, is driven by hidden mental state (what a person believes, wants, intends, feels, and considers socially permissible), so a model that tracks the physical scene but not what each agent knows and believes about it predicts the wrong action for the right-looking scene. We formulate Mental World Modeling (MWM), a generic theoretical framework that makes mental variables core components of a world model rather than posthoc rationales: MWM aintains a coupled physical-mental world state, renders a target-specific partial observation, and simulates how candidate actions jointly update both components. We instantiate the framework in MENTIS, a training-free and fully inspectable baseline that decomposes the process into state parsing, target-observation generation, action decomposition, coupled physical and mental transition, and branch-level value evaluation. On a manually constructed, quality-controlled dataset of situated decision scenarios spanning text, image, and sounding-video stories, experiments with 8 modern LLM-based world models demonstrate that explicitly modeling the mental state is essential for predicting human decisions. Deeper analyses further expose the bottlenecks of current mental world modeling. We expect MWM as a next stage of world modeling, from simulating physical scenes to simulating the minds that act in them.

We present N_0-VTLA, a vision-tactile-language-action (VTLA) foundation model capable of (1) fine-grained contact-rich manipulation with tactile perception and tactile-feedback control, and (2) offline policy improvement from stored deployment data. Building on current vision-based backbones, we propose a training recipe for tactile integration consisting of visuo-tactile pre-training, staged tactile-pathway integration, and advantage-conditioned offline policy improvement. During pre-training, the policy learns broad contact priors from NeoData, our large-scale visuo-tactile robot dataset; to our knowledge, N_0-VTLA is the first VTLA model pretrained on tactile data at scale. During post-training, we augment the policy with a predictive tactile pathway that distills the contact patterns learned at scale into the fine motion adjustments required by downstream tactile-centric manipulation. For offline policy improvement, we introduce ALTER, an advantage-conditioned offline reinforcement learning method that converts relative progress and trajectory-event comparisons into binary advantage labels for policy training on a fixed deployment corpus, further improving task-specific learning on contact-rich skills such as deformable object manipulation. Across contact-rich benchmarks, N_0-VTLA outperforms strong baselines by wide margins: it wins all nine real-robot NeoReal tasks and reaches 63.8% mean success on a twenty-task simulation suite, against 44.0% for the strongest baseline. N_0-VTLA policies trained with ALTER reach 75-95% success on three long-horizon real-robot tasks. These results lay a foundation for versatile tactile-driven manipulation policies.

Polygonal meshes are the standard surface representation of modern 3D pipelines, and generating high-quality meshes with artist-style topology is essential for film, gaming, and interactive 3D applications. Mainstream approaches serialize a mesh into a token sequence and decode it autoregressively, which is slow at inference and sensitive to error accumulation, making them impractical for interactive asset creation. We present Meshy T2, a fast native mesh generation framework built on flow matching. At its core is a vertex-set mesh VAE that encodes a mesh into one continuous latent token per vertex and decodes vertices, edge connectivity, and face winding order in a single pass, preserving high-precision geometry and artist-authored topology without vertex quantization or welding. Generation proceeds as a coarse-to-fine cascade of two flow-matching models: an image-conditioned voxel flow first sketches the overall shape as a coarse occupancy scaffold, and a mesh flow then populates the scaffold with per-vertex latent tokens, conditioned on the image, the scaffold, and a requested vertex budget. This design delivers three practical capabilities: interactive generation speed through parallel flow-based synthesis; effective face-count control through the requested vertex budget; and native support for multi-part assets, whose components emerge directly from the generated connectivity. In our experiments, Meshy T2 achieves state-of-the-art geometric fidelity and completes end-to-end image-to-mesh generation within a median of 6 seconds, over an order of magnitude faster than autoregressive baselines. Code and weights will be available at https://github.com/meshy-dev/meshy-t2.

System prompts are instructions configured by developers to govern the behaviors of foundation models in AI applications. They are used throughout commercial AI products, but are rarely disclosed to the public or regulators, creating a serious trust and accountability gap in the wide deployment of AI systems. In this paper, we introduce Artificial Intelligence System Prompt Assurance (AISPA), a user-centric framework for systematically auditing system prompts in AI systems. AISPA examines specific parts of a system prompt and evaluates them along eight dimensions that matter to users. We then use this framework to review 3,249 instructions from system prompts in 88 commercial AI products, classifying each instruction as either protective (of users) or problematic. Our audit surfaces four core findings. First, system prompt design varies substantially across products and developers, with some organizations averaging over 60 protective instructions per product while others average fewer than 5. Second, protective instructions are widely adopted but shallow in scope: 98.9% of products contain at least one, yet only 24% cover all eight dimensions of the AISPA taxonomy. Third, system prompts have grown steadily longer and more protective of users, suggesting that user protection is becoming a more visible concern in commercial prompt design. Fourth, despite this progress, problematic instructions remain pervasive: roughly 40% of products contain at least one instruction that works against user interests, and protective and problematic instructions frequently coexist within the same prompt. Our findings highlight the need for greater transparency, standardization, and independent oversight for system prompts in commercial AI products.

We present N_0-TWAM, a tactile-native world-action model for contact-rich manipulation that predicts both future vision and future contact. To our knowledge, it is the first tactile world-action model trained at large scale, and it shows strong capability on contact-rich tasks. We pre-train N_0-TWAM at large scale with visuo-tactile joint training over tactile-rich demonstrations spanning six embodiments and 450 tasks. We use NeoForce, a unified force-based tactile representation, to form a physically grounded contact signal that conditions action generation. To improve long-horizon and multi-stage manipulation, we introduce tactile contact events for task staging and advance through them during execution. For real-time efficiency, we adopt an asymmetric Mixture-of-Transformers architecture that pairs a full-width expert for video prediction with slim experts for downstream action and tactile prediction. Evaluations on both real and simulated benchmarks justify the capabilities of N_0-TWAM across a range of contact-rich tasks, and demonstrate the benefit of data scaling for precise tactile and action prediction. In summary, N_0-TWAM endows a world-action model with predictive capabilities to foresee vision, touch and action, building a solid foundation for fine-grained manipulation on open contact-rich tasks. The codebase and model checkpoints will be made publicly available to foster further research and development in tactile-enabled robotic manipulation.

Reinforcement learning with verifiable rewards (RLVR) broadcasts a single response-level reward to every token, while on-policy distillation (OPD) scores each token against a stronger teacher for a dense advantage but caps performance at teacher quality and discourages exploration beyond it. Their complementarity makes combining RLVR and OPD promising, but we find that fusing the two advantages with a fixed coefficient triggers entropy collapse from two miscalibrations: a magnitude mismatch, where token-level OPD advantages can spike far beyond the bounded RLVR advantage and erase its signal, and a temporal mismatch, where sustained full-strength OPD keeps pulling the student toward the teacher and limits exploration needed to surpass it. We propose SAF, a Stable Advantage Fusion framework that resolves both issues via a lightweight, four-stage pipeline applied only to the OPD advantage: a sparsify-then-compress mechanism for magnitude control paired with a warm-up-then-anneal mechanism for temporal control, with each stage independently switchable and adding negligible overhead. Instantiating RLVR with GRPO, we evaluate SAF across seven mathematical reasoning and code generation benchmarks with Qwen3-1.7B/4B/8B: SAF avoids entropy collapse and consistently outperforms fixed-coefficient GRPO+OPD fusion, improving the aggregate score by 0.51-2.70% across all six model-domain settings while achieving more stable training.

We study empirical scaling properties for text conditioning in visual generation. Such properties have rarely been measured because diffusion loss does not scale with the number of tokens in natural-language prompts. Surprisingly, we find that the converged diffusion loss scales with the amount of structured language in the prompt. To quantify structured language, we adapt two complementary measures: a white-box likelihood metric (GPG) and a black-box attribute metric (ED). Across controlled training runs, the converged diffusion loss decreases approximately linearly with GPG and follows a power law with ED. Guided by these scaling properties, we improve diffusability by constructing structured prompts with semantic and geometric annotations derived from images, and improve promptability by training a prompter through supervised fine-tuning, cold-start, and verifier-gated on-policy distillation. The resulting system outperforms all evaluated open-weight models on nearly every compositional, reasoning, and world-knowledge benchmark, while matching or surpassing the strongest closed-weight models on most evaluations.

Latent world models enable efficient planning by predicting future states in a compact representation space, but their performance depends critically on the quality of the learned latent distribution. LeWorldModel (LeWM) regularizes its latents toward an isotropic Gaussian using the Epps-Pulley (EP) objective. We show that the corrective gradients of EP rapidly vanish for isolated tail samples, leaving heavy-tailed deviations insufficiently controlled. To address this limitation, we propose QQWorld, which replaces EP with a quantile-quantile matching objective that directly aligns projected latent samples with rank-matched Gaussian quantiles, thereby maintaining effective corrective gradients in the tails. We further develop cross-batch QQ, which enlarges the effective ranking pool using detached samples from previous batches, and characterize its bias-variance trade-off. Across four control environments, QQWorld effectively improves the average planning success rate of LeWM, while consistently yielding better Gaussian alignment and thinner latent tails.

AI-assisted coding increasingly translates informal user intent into executable software, yet coding requests often contain ambiguities that recur in user-specific ways across tasks and sessions. Existing disambiguation methods typically address each ambiguous request in isolation within the current coding session, often through eliciting additional clarification. However, whether resolved session history from the same user can serve as memory for resolving recurring personalized ambiguity in a newly opened session remains underexplored. We formulate personalized ambiguity adaptation as a new task: given a user's previously resolved coding sessions and a new ambiguous request, an assistant should identify the recurring ambiguity pattern, produce the intended executable solution, and minimize clarification. To benchmark this task, we introduce CAPA, which characterizes personalized coding ambiguity through six mechanisms and injects these mechanisms into unambiguous executable tasks using a controlled three-stage generation pipeline. CAPA contains 600 coding sessions across 60 balanced user--ambiguity cells, including 300 held-out evaluation sessions. We evaluate 12 recent LLMs under no-history and same-user-history conditions using executable success, first-turn success, and turns-to-completion. Our analyses examine task difficulty, user identity, and memory-based history use, and we further propose same-user history gating as a lightweight inference-time method. CAPA provides a foundation for developing long-term coding assistants that better align generated code with user intent while reducing repeated clarification.

Rubric-based RL has recently shown promise in improving LLMs on open-ended tasks. A widely recognized limitation of rubric-based RL is limited exploration: criteria that no rollout manages to satisfy (Unexplored Criteria, UC) receive no optimization signal. Recent methods address this by incorporating rubric information as external guidance during rollout, yet they introduce a train-inference mismatch: the policy is optimized on rollouts produced under external guidance while this guidance is absent at inference time, causing error accumulation through autoregressive decoding. Moreover, these exploration-focused approaches overlook a fundamentally different failure mode that we term Suppressed Criteria (SC) -- criteria that are satisfied by some rollouts yet whose learning signals are lost during optimization because scalar reward aggregation assigns them non-positive aggregate advantages. Our analysis reveals that SC are remarkably prevalent: over 57% of samples exhibit this failure mode throughout training, with an average of 1.8 SC per sample. To simultaneously address both UC and SC without introducing training-inference mismatch, we propose Criterion-Distilled Policy Optimization (CriPO), which enhances rubric-based RL via on-policy self-distillation. For UC, CriPO constructs a criterion-injection self-teacher and computes a localized forward-KL loss to inject missing behaviors into the policy. For SC, CriPO employs a counterfactual self-teacher to locate criterion-relevant tokens in negative-advantage rollouts and flips their token-level advantages to positive values, preserving useful patterns that would otherwise be suppressed. Experiments on medicine and science benchmarks demonstrate that CriPO consistently outperforms rubric-based RL, achieving stronger final performance with approximately 2times fewer optimization steps.

Enterprise workflows increasingly rely on agents for schema-guided extraction: given a document and a user-defined schema, the agent faithfully follows the schema to produce the correct output with source evidence as grounding metadata. We present ExtractBench, a benchmark for schema-guided extraction and, to our knowledge, the first to score value accuracy, record completeness at scale, grounding, and measured cost together. The evaluation system contains 4,869 pages across 370 enterprise documents, 8 business domains, and 67 document types, with clear tags differentiating their challenge scenarios. The scalable schema and ground-truth curation pipeline combines independent-system agreement for real documents, known values for synthetic lists, and human verification for forms. We report order-insensitive value F1 for value accuracy, plus two grounding metrics for source traceability: word- and page-level F1. Commercial VLMs perform well on short documents but often truncate record lists on long ones, while coding agents retain higher accuracy at much higher cost. LlamaExtract Agentic Plus ranks first on all three metrics, with accuracy comparable to coding agents at a fraction of the cost. Dataset and evaluation code are available on https://huggingface.co/datasets/llamaindex/ExtractBench{HuggingFace} and https://github.com/run-llama/ExtractBench{GitHub}.

While recent image editing models have made rapid progress, multi-reference editing remains challenging, particularly in maintaining visual consistency across references and ensuring overall visual harmony. Reinforcement learning has proven highly effective for text-to-image generation and single-image editing, but its extension to multi-reference editing is hindered by the absence of suitable reward models that capture multi-image relational constraints. Moreover, naively using multimodal large language models(MLLMs) as zero-shot evaluators faces a key tension between hallucination-prone long-form reasoning and the limited deductive power of short-form judgments. We address these issues with a Multi-dimensional Evaluation-Verification Reward(EVR). EVR decomposes evaluation into distinct visual criteria; for each criterion, an MLLM Evaluator generates multiple candidate hypotheses, and a Verifier grounds each claim in concrete visual evidence to accept or reject it, producing reliable and fine-grained reward signals. Together with a scalable data pipeline, our method enables RL fine-tuning of off-the-shelf editors without architectural changes. Extensive experiments show substantial gains over the base Qwen-Image-Edit, improving consistency and harmony to match or surpass NanoBanana.

In the physical world we inhabit, space and time are fundamentally continuous. However, existing machine learning paradigms for world modeling are largely confined to discrete-time prediction, thereby exhibiting significant inefficiency in capturing the dynamics of physical world. We introduce Physical-Time Flow (PT-Flow), a novel approach that learns a continuous latent velocity field operating in physical time. Crucially, the underlying dynamics of sequential data are parameterized by an ordinary differential equation (ODE) embedded in a well-structured representation space. Under this paradigm, the prediction of future can be recast as temporal integration via an ODE solver in the compressed latent space. Building upon PT-Flow, we construct ODEWorld, a continuous-time latent world model that is both efficient and versatile. By extracting time-variant features and enforcing ODE properties on both the dynamical representation space and the latent velocity field, ODEWorld effectively addresses the long-standing representation collapse issue in latent world model literature. This also enables high-quality image reconstruction even after long-horizon prediction. Moreover, its continuous nature allows for arbitrary temporal resolution and even backward prediction, which is impossible for most discrete-time models. Lastly, ODEWorld can provide rich planning-oriented information to facilitate downstream policy learning. Comprehensive experiments demonstrate that ODEWorld successfully reconciles planning-conducive dynamics abstraction with visual realism, excelling in both video generation and robotic control. https://dstate.github.io/odeworld_website/{Project Website}.

The web is increasingly accessed by AI agents rather than humans. Every agent needs knowledge, especially in the life-sciences, where agentic pipelines are growing fast. Access to the literature is a crucial part of that need, and resources such as Europe PMC, with over 40M indexed records, are widely used to meet it. Yet these resources were not built for AI agents: they take keywords and complex syntax and return whole papers, so every agent must learn the syntax, issue several searches, and read full papers to find the evidence it needs. We introduce EMBL AI Librarian, a knowledge layer that upgrades the Europe PMC interface for AI agents: an agent asks in natural language and receives evidence that answers it. A single LLM orchestrates the whole knowledge retrieval process: it plans complementary subqueries executed by the live Europe PMC search engine, then reads the selected papers and locates the relevant evidence. We evaluate Librarian across four benchmarks: literature synthesis, claim verification, open-domain question answering, and downstream biology tasks such as protocol questions and sequence manipulation. On ScholarQABench, Librarian improves Citation F1 by more than 16 points over strong recently published baselines. Used as the retrieval layer of an existing claim-verification pipeline, it increases agreement with expert consensus; and on the open-form LitQA2 benchmark, a GPT-5.4 agent scores about 8 points higher when grounded in Librarian than with web search. Overall, our results show that equipping life-science agents with the Librarian knowledge layer improves performance across a range of tasks. We release our code publicly at https://github.com/petroni-lab/librarian

Jeran Wittenstein / Bloomberg : Amazon becomes the fifth company to reach a $3T market cap, joining Nvidia, Alphabet, Microsoft, and Apple; its stock closed up 4.58% on Monday —  Amazon.com Inc. surpassed $3 trillion in market value for the first time, becoming only the fifth company to ever reach the milestone.

Jonathan Vanian / CNBC : Snap reports Q2 revenue up 19% YoY to $1.6B, vs. $1.54B est., DAUs up 5% to 493M, vs. 487M est., forecasts Q3 revenue above est.; SNAP jumps 10%+ after hours —  Snap reported better-than-expected revenue and earnings for the second quarter and issued a forecast for the current period that topped analysts' estimates.

Samantha Subin / CNBC : Palantir reports Q2 revenue up 93% YoY to $1.94B, vs. $1.8B est., US commercial revenue up 149%, and raises FY 2026 revenue guidance; PLTR jumps 9%+ after hours —  Palantir topped second-quarter earnings estimates on Monday and said commercial revenue more than doubled from a year ago.

Zack Whittaker / TechCrunch : Samsung says it's restricting app registrations and removing apps with SDKs that can turn smart TVs into residential proxy exit nodes, after a similar ban by LG —  Several popular Samsung smart TV apps contain code that share the owner's internet connection with strangers …

Devlin Barrett / New York Times : Court docs: an FBI agent has been charged with amassing ~$1M worth of cryptocurrency, largely by making unauthorized withdrawals from a criminal target overseas —  The agent confessed to making unauthorized withdrawals from a criminal target overseas, according to court papers.

Michael Kan / PCMag : US satellite internet provider Hughesnet files for Chapter 11 after losing subscribers to Starlink and plans to focus on business and government customers —  US satellite internet provider Hughesnet has filed for Chapter 11 bankruptcy after running low on cash and losing subscribers to Starlink.

The Verge : Leaked doc: Microsoft plans to bring Xbox 360 games to PC, and the games will be able to run on its next-gen Project Helix console, “Xbox PCs”, and handhelds —  The disc to digital program seems like it's close to launching, too.

Maria Curi / Axios : The White House says it has met its deadline to establish a voluntary framework for evaluating advanced AI models; it did not provide details of the framework —  The White House said on Monday it met its deadline to establish a voluntary framework for evaluating advanced AI models …

Eduardo Baptista / Reuters : China revises its regulations to tighten protection of chip layout designs, including punitive damages for serious infringement, taking effect from October 15 —  China has revised regulations protecting integrated-circuit layout designs, tightening registration standards and allowing punitive damages …

Bloomberg : Valar, which is making small modular nuclear reactors to power data centers, raised a $1B Series B led by Sequoia at a $6B post-money valuation —  Sequoia Capital led a $1 billion funding round for Valar Atomics Inc. that the nuclear startup says will help it shift from demonstrating small reactors to producing them in volume.

Allie Garfinkle / Fortune : Mariana Minerals, which builds and operates mines using its MarianaOS software platform, raised a $310M Series B, bringing its total funding to $400M —  The last century was fueled by oil and gas.  Turner Caldwell is betting the next century runs on metals.

Reuters : India proposes extending tax exemptions until 2041 for foreign companies that provide machinery to contract manufacturers, changes that Apple had lobbied for —  India has proposed extending tax exemptions until 2041 for foreign companies that provide machinery to contract manufacturers …

The Information : Sources: the Trump administration invites staffers from OpenAI, Google, Anthropic, and others to the White House on Tuesday to review the AI oversight framework —  The Trump administration has invited staffers from major tech companies including OpenAI, Google and Anthropic to the White House …

Tim Bradshaw / Financial Times : UK court filing: in July, Apple launched a new legal challenge against the UK government's attempt to create a “backdoor” to access encrypted customer data —  iPhone maker files complaint over latest government demand to allow access to encrypted user data

Meir Orbach / CTech : Zenity, which develops a platform for securing AI agents, raised a $125M Series C led by Norwest Venture Partners, taking its total funding to ~$185M —  Israeli cybersecurity company reaches $185 million in total funding as enterprises race to secure autonomous AI agents.

微软 Windows 和设备执行副总裁 Pavan Davuluri 通过官方博客谈论了该公司在 Windows 11 上持续展开的改进工作，虽然大部分工作尚未推送给用户。他透露接下来的一项工作是优化 Windows 11 在 8GB 内存以上配置下的运行效率。由于目前的内存供应危机，PC 厂商又开始提供 8GB 内存配置的电脑，然而 Windows 11 在 8GB 内存下只能勉强运行，运行其它程序面临很大的局限性。Davuluri 表示该公司将减少 Windows 的内存占用，在用户日常使用的 PC 上提供快速响应的 Windows 体验。

塑料颗粒显然已经遍及地球的各个角落。根据发表在《Water Research》期刊上的一项研究，韩国研究人员检查了生活在水下 2000 米处深海热泉附近的蜗牛和贻贝，12 只受检动物 92%（11 只）体内都发现了微小的塑料碎片。海洋生物学家 Se-Joo Kim 说，深海热泉曾被认为是地球上最与世隔绝的环境之一。每只深海热泉动物体内平均发现了 3.42 个塑料碎片，最常见的是聚苯乙烯。研究人员还发现，摄食行为会影响微塑料在动物体内的积累位置。海床进食的蜗牛，塑料聚集在消化器官中；滤食性贻贝的塑料分布则更为均匀。

澳大利亚联邦银行、微软、 Uber 和凯悦酒店等公司正用自动化聊天和电话系统去处理以前需要人工完成的工作。AI 系统开始大幅削减客服人数。呼叫中心行业在美国、印度和菲律宾等地雇佣了数百万从业者。随着 AI 技术的进步以及高管面临拥抱新技术的压力，企业日益用生成式 AI 工具取代客服。美国的客服务人数正在下降，未来这一趋势会继续下去。分析师估计到 2030 年近半客服岗位将受到影响。西方公司此前将客服外包到熟悉英语的菲律宾、印度等国，随着 AI 的普及，这些国家的客服人数预计会大量减少。澳大利亚最大银行——联邦银行（Commonwealth Bank of Australia）裁减了数百名客服，每年预计节省数千万美元。微软将其客服团队（其中包括合同工和全职员工）从约 5 万人缩减至 4 万人。微软销售和服务运营负责人 Judson Althoff 在 4 月曾表示，AI 每年为公司节省约 7.5 亿美元的客服成本。他表示，更复杂的问题仍需要人工支持，但公司正不断扩展自动化修复的范围。凯悦酒店去年裁掉了美洲地区三成的内部客服。Uber 裁减了 10% 的客服岗位以“拥抱 AI”。

澳大利亚政府为 16 岁以下儿童的社交媒体禁令辩护，称禁令重塑了全国的舆论，极大改变了与家长之间的对话，关闭了数百万​​个儿童账户。政府从未指望禁令能得到 100% 的遵守，就像最低饮酒年龄法无法做到 100% 遵守一样，但法律的存在仍然是恰当的。此前的研究发现，禁令生效三个月后，逾八成澳大利亚青少年仍在继续使用社交媒体，10-15 岁儿童在 3 月使用社交媒体的频率与 12 月 10 日禁令生效前一样高。

在 Linux 游戏社区多年请求之后，GOG Galaxy 游戏启动器将原生支持 Linux。GOG 有游戏的 Linux 版本可以下载，但只能通过网站下载，GOG Galaxy 客户端只支持 Windows 和 macOS 操作系统。GOG 没有公布支持 Linux 的时间，只是表示正在开发中。GOG Galaxy 类似 Steam 客户端，支持自动游戏安装和更新、云端存档同步、成就追踪、游戏时长统计、游戏库管理、商店浏览、 好友列表和社交功能等等。在 Valve 的 Proton 推动下，Linux 对游戏的支持日益改进，越来越多的游戏公司也认识到了 Linux 的重要性。

加州居民从 8 月 1 日起可要求数据经纪商删除个人数据。Delete Request and Opt-Out Platform (DROP) 允许加州居民通过一次性选择加入的方式，从数百个数据处理中心删除个人信息。此前用户要删除个人信息需要联系每个数据处理商。数据处理商需要在 90 天内删除用户的数据，之后每 45 天删除一次。加州是美国第一个创建易于使用的数据删除系统的州，

硬件钱包本应是保存比特币最安全的地方：不联网、密钥不离设备，攻击者需物理接触才能盗取加密货币。Coinkite 制造的比特币硬件钱包 Coldcard 使用硬件随机数生成器产生 128 位熵值，这个数字是如此之大现有的计算能力无法破解。然而 2021 年 3 月的一次更新引入了严重 bug，默认的硬件随机数生成器被切换为软件随机数生成器，在 Mk3 设备上，有效搜索空间缩小到 40 位。128 位和 40 位之间的差异是无法被打开的锁和可以被打开的锁之间的区别。攻击者可以租用云计算资源进行暴力破解，在无需接触硬件的情况下获得钱包的密钥。过去几天，攻击者从 4585 个地址窃取了 1367 个比特币，价值近 8900 万美元。Coinkite 已释出新固件，受影响用户需在更新后的硬件上生成全新钱包，将比特币迁移至新钱包。

马来西亚柔佛州警方上个月打击了加密货币矿工的偷电行为，逮捕了三名嫌疑人，缴获了电脑、路由器、汽车以及比特币矿机。该团伙绕过电表偷电，导致电力公司一个月内损失约 14,500 欧元，而比特币矿机每月可产生17,200-21,500 欧元的收入。这起案件价值相对较小，但 2020-2025 年马来西亚能源公司识别了近 14,000 处场所涉及偷电挖加密货币，累计损失 11 亿欧元。加密货币挖矿偷电事件从 2018 年的 610 起增至 2024 年的 2,397 起。加密货币挖矿本身并非犯罪行为，但政府发现非法加密货币挖矿与网络赌博、洗钱以及东南亚大规模电诈之间存在关联。

AI 真的会推理吗？它展示的推理链/思维链真的是导出结论前的逻辑步骤？或者它给出的推理过程只是在糊弄我们？一种拟人化的虚构？Quantamagazine 上的一篇文章讨论了这一问题，作者认为 LRM（large reasoning models）在解复杂问题时常常能给出正确答案，但整个机制仍然是不透明的，是一种黑盒子，而思维链更像是某种自言自语而不是严谨的推理。AI 专家 Melanie Mitchell 指出，LRM 有效，执行推理任务的准确率比 LLM 高；思维链不一定忠实反映模型内部的运作过程；模型合成的思维链文本基本无用完全可以删除。2025 年美国东北大学和加州伯克利的研究人员分析了开源前沿 LRM 模型，发现 30%-60% 的思考步骤对模型在数学题上给出的答案“几乎没有因果影响”。

韦伯望远镜（JWST）正以前所未有的红外线敏感度揭开宇宙面纱，除了探究早期宇宙，更开启了系外行星大气研究的新时代，而在所有观测目标中，最吸引人的莫过于生命不可或缺的关键要素——水。天文学家利用韦伯望远镜的数据，对距离地球约 700 光年的系外行星 WASP-39b 进行大气模型分析，并首度在其大气中侦测到了半重水（semi-heavy water）的存在。这项研究重点在于寻找水的同位素异构物，半重水与一般水分子 H2O 的差别在于其两个氢原子中，其中之一被带有一个中子的氘同位素所取代。研究人员希望透过精确测量大气中水分子的氘与氢（D/H）比例，推算这颗行星的诞生与演化历程。

维基媒体基金会于 7 月 27 日以美国科技公司的典型做法拒绝自愿认可员工工会，声明中充斥着典型的反工会言论，声称它的 342 名美国员工对加入工会持有不同的看法，它坚持要求由美国国家劳工关系委员会（NLRB）裁决维基员工组织工会的权利，其目的被认为是拖延进程、阻挠投票，以便让反工会人士有时间对基金会员工发起可能具有强制性的反工会宣传。根据 NLRB 网站上的一份文件，维基媒体基金会雇佣了 Littler Mendelson 律所作为其法律代表，该律所以反工会臭名昭著。

Valve 公布了 2026 年 7 月的 Steam 硬件和软件调查，Steam 用户中使用 Linux 比例从 3 月的 5.33%、4 月的 4.52%、5 月的 3.99% 以及 6 月的 3.69% 之后再次超过 4% 达到 4.01%。Windows 操作系统的比例占 93.67%，OSX 占 2.32%，其中 Windows 11 占 70.26% 而已经停止支持的 Windows 10 占 23.30%，游戏玩家中仍然有超过五分之一使用 Windows 10。其它数据显示：简体中文用户比例占 22.52%，英语用户占 39.61%。玩家使用英特尔 CPU 的比例占 53.76%，AMD 占 46.24%。

腾讯旗下工作室 Lightspeed LA 本周宣布裁员 80 人，调整其开放世界游戏《最后的哨兵（Last Sentinel）》的创意和开发方向。开发历时六年，多次延期，预算高达数亿美元的《最后的哨兵》未来前景不明。《最后的哨兵》是腾讯投资最高的 3A 游戏，也是 Lightspeed 的第一款游戏，意图挑战 Rockstar 的侠盗猎车手系列，招募了前 Rockstar 的资深开发者 Steve Martin 领导工作室。由于对游戏现状不满，腾讯给工作室下了最后通牒，要求在 7 月底前开发出一个更好的版本。在截止日期前 Martin 等人去腾讯总部展示了新版本。结果是宣布了裁员。目前不清楚腾讯会如何处理《最后的哨兵》，可能会推出一个精简版本，也可能会将现有素材用于其它项目。

根据 Pew Research Center 的调查，77% 的美国成年人支持初中生和高中学生在课堂上禁止使用手机，这是支持率连续第二年上升——2024 年这一比例为 68%。只有 18% 的人反对禁令，6% 的人表示不确定。48% 的美国成年人支持初中生和高中学生在整个学校期间都禁止使用手机，前一年的比例是 36%。但仍然有 43% 的人反对该禁令。18-29 岁年轻人对课堂（57%）和全天（28%）手机禁令的支持率最低，65 岁及以上人群对禁令的支持率最高。共和党人更倾向于支持课堂和全天手机禁令，而民主党人则更支持课堂禁令而非全天禁令。美国已有 42 个州制定了在学校期间限制使用手机的政策，但有的限于课堂，有的则禁止全天使用。

记者 Gil Duran 即将出版的新书《The Nerd Reich: Silicon Valley Fascism and the War on Democracy》探讨了硅谷科技寡头的企业帝国之梦。亿万富翁们对社会有两种态度。其一称之为“退出”，他们想要去其它地方创建一个新的国家，认为民主制度太强大，太混乱，反弹太强烈，这是提出“The Network State”的 Balaji Srinivasan 等人的立场。其二是利用其富可敌国的财富改变和控制社会，这就是硅谷亿万富翁们现在正在做的。2024 年 9 月举行的名为 Reboot 2024 的会议讨论了右翼特朗普支持者和硅谷风险投资家之间的合作，出席者包括了 Peter Thiel 的副手 Michael Kratsios、传统基金会的 Kevin Roberts、Y Combinator CEO 陈嘉兴（Garry Tan）。Curtis Yarvin 被视为是这场运动的奠基人之一，他曾撰文论述用由 CEO 统治国家的制度去取代美国的民主制度。Thiel 以及 Marc Andreessen 都对 Yarvin 推崇备至，经常引用 Yarvin 的评论。亿万富翁中 Elon Musk 可能吸引了最多的眼球，但幕后真正的掌权者是 Peter Thiel，他的 15 名副手在特朗普政府高居要职。副总统 JD Vance 的整个职业生涯都是由 Thiel 一手提拔的，他距离总统之位一步之遥。金钱集中在少数人手中，他们拥有过大的权力，而这些人皆非心胸开阔之辈。这群亿万富翁的主要创新是不再承诺回归神话般的过去，而是承诺一个神话般的未来：超级富足、星际旅行、永生、极致效率。