OrangeBot.AI Digest — 2026-08-18

Local AI dictation for Windows

Build deterministic agents that actually follow your process

Your AI coworker in Slack

Borrow the best AI setups and make them yours.

Create and manage ElevenLabs voice agents in your chat

Chat to yourself and find anything by what you remember

Easily manage devices and switch them between Macs

Search your bookmarks by what's inside them, not the title

Run AI locally and own the whole workflow

Build Smarter Agents using Structured Context

A persistent terminal for Codex and Claude side by side

Turn website visitors into qualified pipeline

Create a plan for anything. Share it with anyone.

Agent Led Growth: Get written into every customer's codebase

Funnel analytics you'll actually understand

A benchmark should deliver more than a scalar score: what makes an evaluation trustworthy is the reasoning that justifies the score. This is especially critical for world models, where judging a rollout requires understanding whether physics, causality, and world state evolve correctly. Humans spot such violations naturally, yet no existing benchmark automates this capability: metrics are computed brute-force, leaving no reasoning chain that can be examined or verified. We introduce HarnessEval-W, an agentified evaluation pipeline that brings the harness paradigm from the LLM ecosystem to world model benchmarking. Rather than applying a fixed rubric, HarnessEval-W interprets the context of each evaluation case, decomposes the evaluation question into measurable subproblems, and spawns specialized sub-agents, each equipped with tailored context and diagnostic tools to reason over its own subproblem. The parent agent then validates the gathered evidence and summarizes it into the final verdict. This hierarchical workflow turns every evaluation into a transparent evidence tree whose complete reasoning chain justifies the result. We apply HarnessEval-W to 18 representative world models over 330 evaluation cases. Its judgments closely align with human preferences while providing verifiable, fine-grained diagnoses of every generated rollout. We open-source the full pipeline as a live benchmark and invite the broad community to contribute to grow new skills and evaluation cases as world models evolve.

Long-horizon agents can fail even when their underlying models can solve the constituent steps. They may lose track of mutable state, fail to reactivate lessons from earlier executions, skip known procedures, or stop prematurely. We bet on harness scaling to improve the execution system around an agent without changing its model weights. We introduce StateM, an agent-native runtime that organizes execution around durable states, phase-local context, checked transitions, recoverable runbooks, and versioned procedural practices that agents and users can inspect together. On Terminal-Bench 2.1, StateM raises GPT-5.5 xhigh to 92.1\%, versus 83.1\% reference and GPT-5.6 Sol Ultra at 91.9\%. The runbook transfers unchanged to GPT-5.6. With GPT-5.6 Sol xhigh, StateM reaches 95.3\% raw accuracy across 445 trials and succeeds on all 89 tasks at least once. The frozen profile raises GPT-5.6 Luna from 76.7 to 85.4\%, above the 84.9\% Sol xhigh reference. Using the same runtime, runbook structure, and golden rules, less than \38 of adaptation raises DeepSeek-V4 Flash from 82.7 to 88.1% under standard timeouts and to 89.1% on an 88-task common core. Extending only the remaining latency-sensitive task matches the reported 88.8% GPT-5.6 Sol max result. Final-score API usage is about 15 versus \574.68 for the GPT reference; total DeepSeek expenditure is 52.22. On BusinessBench, family-specific runbooks built on development sets yield held-out gains of 0.55 macro and 1.34 micro points; two mechanism-matched families improve by 10.04 points. Concrete rules generalize when tasks share execution structure, while the control methodology applies broadly. StateM turns selected postmortem findings into persistent, executable preconditions and practices, making learned controls explicit and enforceable through stateful controls. Code at github.com/henryqin1997/statem.

Constructing an interactive 3D open world from a user query is important. However, existing methods are primarily evaluated on idealized, simple queries, making it difficult to systematically analyze and compare how multimodal agents understand user intent, use 3D tools, and reason over textual and visual 3D world information. To this end, we propose VibeWorlding, a unified framework for benchmarking and training vibe worlding agents: a multimodal agent that can autonomously infer user intent, plan scene layout, invoke 3D tools, and reflect on the multimodal feedback in a multi-turn agent-environment interaction process. To achieve this, we first build VWE-BENCH, a benchmark of 2,616 high-quality 3D assets, 323 human-annotated seed 3D worlds, and 6,828 reverse-synthesized multimodal user queries, split into verified queries with ground-truth and unverified queries with carefully designed rubrics. Moreover, we develop VibeWorlding-Gym, a joint multimodal RL post-training framework that integrates (1) a sandbox environment unifying asset retrieval, editing, and image rendering as MCP tools, and (2) a rubric-based verifier that combines physical feasibility and intent fulfillment verification, supporting both fair model evaluation and scalable multimodal RL reward service. Our experiments show that current frontier MLLMs are far from solving the vibe worlding agent task, with even GPT-5.5 and Qwen3.8-Max reaching below 60% success rate, and trace the bottleneck to precise 3D world editing. We further find that RL training can ease this weakness and enable open-source MLLMs to even surpass closed-source frontiers: our VibeWorlder-8B is comparable to frontier MLLMs, while our flagship VibeWorlder-30B-A3B attains the best overall Pass@1 among all evaluated models.

Scientific discovery often involves optimising expensive-to-evaluate objectives over vast, structured, and open-ended hypothesis spaces, such as molecules, protein sequences, and computer programs. Generative models such as large language models (LLMs) provide expressive priors over such spaces, but their likelihoods and self-assessments are unreliable proxies for the objectives and calibrated epistemic uncertainty, especially for novel candidates outside the observed data distribution. We introduce the Large Discovery Model (LDM), an empirically grounded recurrent architecture that couples a generative model with a Bayesian non-parametric reward surrogate model. The generative model proposes and refines candidate designs, while the surrogate predicts their performance and quantifies uncertainty, yielding an uncertainty-aware value that guides candidate generation, refinement, and selection. The discovery memory and the surrogate model are continually updated as each new experimental observation arrives. We evaluate LDM on three scenarios spanning different design modalities and objectives, including neural-network training, antibody design, and molecular optimisation. Compared to LLM-only reflection or traditional statistical search across these domains, LDM achieves a 2.4times greater reduction in validation BPB, an 18.2% relative decrease in binding energy, and more than 60% relative gains in molecular multi-objective performance. These results suggests that LDM could serve as a general-purpose discovery engine for effective search over open-ended hypothesis spaces.

Reinforcement learning (RL) with group-relative advantages has become the de facto standard for post-training language model reasoners. However, when optimizing multiple reward objectives, existing methods typically scalarize the reward vector with a fixed weighted sum before group-wise standardization. We show that this design leads to two fundamental problems: rollouts with distinct reward profiles can receive identical advantages, and all objectives are optimized with fixed relative weights regardless of their current level of saturation. As a result, training continues to allocate gradient budget to already-solved objectives instead of focusing on those with greater remaining headroom. We introduce Saturation Aware Advantage Reweighting for Multi-Reward Policy Optimization (SA-MRPO), which standardizes each reward objective independently and adaptively discounts its contribution according to a batch-level estimate of objective saturation. This dynamically reallocates optimization effort toward under-optimized objectives while empirically maintaining performance on those that are already well satisfied. We further show that saturation-aware reweighting can reverse the sign of an update, rather than merely rescale its magnitude. Across mathematical reasoning with two- and three-objective reward combinations, SA-MRPO improves the harder correctness objective over GDPO in 12 of 15 benchmark comparisons, with gains of up to 5% on AIME24. On adaptive reasoning it improves accuracy on all five benchmarks, by 3.8% on average and up to 9.2 % on AMC23, and on coding benchmarks it improves pass rate by up to 2.3%, while in all settings maintaining the easier objectives near their already satisfied levels.

We present MOSS-VL, an open vision-language model family that treats real-time interaction -- perceiving while it speaks -- as a first-class capability. It is co-designed across the stack: the language decoder attends to vision only through gated cross-attention, so the model can naturally see incoming frames while generating; a synthesized interaction corpus supervises when to speak, when to stay silent, and when to revise; and a staged curriculum concentrates all real-time-specific training in one light final stage over a strong offline foundation. Offline, MOSS-VL-Instruct is competitive at comparable scale and leads temporal-reasoning video sets. Across four streaming benchmarks, MOSS-VL-Realtime posts the best average on three (second on the fourth) among open-source streaming models, sweeping the three subsets that squarely test proactive behavior -- 66.0 vs. 37.5 for the best baseline on OmniMMI Proactive Alerting. With 11.3B parameters but visual tokens outside the decoded sequence, MOSS-VL widens its time-to-first-token advantage over same-backbone Qwen3-VL-8B from 2.8x to 5.1x as visual context grows. We release all five checkpoints, the training curriculum, and the real-time inference code at https://github.com/OpenMOSS/MOSS-VL.

Foundation GUI agents can automate complex digital tasks, but deployment is hindered by scarce and biased training data, ambiguous prompts, and unreliable execution. Routine workflows rely on user-specific tools and tacit conventions, so unstated instructions can produce arbitrary variations across runs. We present UI-Mate, a foundation GUI agent that integrates an environment-grounded training stack with in-context demonstration learning. UI-Mate makes three contributions: A Scalable Environment-Grounded Training Stack: A closed-loop data engine automates task generation, environment construction, rollout, filtering, capability balancing, SFT, and online RL across massively parallel environments via unified task-verifier bundles. In-Context Demonstration Learning: A mechanism that transforms multimodal demonstrations into flexible subtask-level workflows, follows relevant demonstrated steps, and re-plans from the live interface. OSWorkerBench Benchmark and Insights: A benchmark of 100 long-horizon office tasks across 41 applications that supports instruction-only and demonstration-guided evaluation. Its demonstration resources separate a 33-task self-demo setting, built from successful strong-agent rollouts of the same targets, from a 45-task variant-demo setting, built from human recordings of related but non-identical tasks. Experiments show that UI-Mate-27B sets a new open-weight state of the art on general computer-use benchmarks, scoring 77.0% on OSWorld-Verified and 66.2% on WindowsAgentArena. On OSWorkerBench, it reaches 41.0% strict success and 76.9% progress, outperforming its Qwen3.6-27B base by 17.7 and 24.5 points. On the 33-task self-demo subset, one demonstration raises strict success from 17.2% to 35.4% and progress from 67.9% to 81.1%, substantially improving long-horizon reliability. Project page: https://ui-mate.github.io.

Agent harnesses have substantially improved performance on long-horizon tasks by coordinating agent interactions with the environment. However, reinforcement learning through complex harnesses remains largely unexplored, as scaling such training to long-horizon agent tasks introduces fundamental challenges. In this work, we present a unified black-box RL framework for stable and scalable optimization of general agents through complex harnesses. Concretely, we first build a sandbox-based execution infrastructure that isolates task environments and harnesses within temporary sandboxes for large-scale concurrent rollouts. We then decouple policy optimization from opaque harness execution and place a serving proxy at the model boundary to capture model calls. To reconstruct multi-turn trajectories and improve training efficiency, we organize the captured calls into prefix trees and further adapt both critic-based PPO and critic-free GRPO to optimize over the recovered tree structure. Meanwhile, we maintain training-inference consistency throughout the optimization process. Finally, we introduce mix-harness training, allowing a single model to be jointly optimized by heterogeneous harnesses. With Qwen3-30A3B, black-box RL improves Pass@1 on ClawGym-Bench by 9.98 and 14.81 points through OpenClaw and Claude Code, respectively, while remaining stable over 200-400 optimization steps. Moreover, the framework yields consistent gains on more challenging tasks such as JobBench and OfficeQA. Overall, our framework enables effective, stable, and scalable optimization of general agents through black-box harnesses, supporting unified training across heterogeneous execution systems.

This paper investigates an increasingly important topic in generative modeling: pixel-space diffusion models. Although numerous studies have explored this topic, most focus on small-scale or class-conditional settings. Consequently, a practical recipe for training pixel-space models that rival or exceed well-established latent-space counterparts remains elusive. Through a comprehensive empirical study, we first observe that direct large-scale pre-training in pixel space converges substantially more slowly than in latent space. This observation motivates a latent-to-pixel strategy that acquires generative priors efficiently in latent space and transitions to pixel space during post-training. We then systematically investigate the key design choices governing this transition, including weight initialization, data composition, prediction target, decoder architecture, and noise schedule, and identify a practical recipe that makes the resulting pixel-space models match or outperform their latent-space counterparts while delivering 3.18 to 4.75 times end-to-end inference speedups. We hope that our findings provide useful empirical insights and practical guidelines for future research on pixel-space generation.

This first release of Prior Labs in relational learning shows our continued commitment to open science. We open-source three pieces of software that we expect to accelerate research in the field towards meaningful real-world impact. We aim to steer further development based on feedback from, and in collaboration with, the community. Given the early stage of development, our α-release targets researchers and early-adopting practitioners. Over the past years, a variety of datasets and tasks for relational learning have emerged, but the community has not converged on a reliable, reproducible way to compare different methods on these tasks. Our α-release, RelArena-α, provides a unified framework for running and comparing baselines on RelBench v1 by standardizing data loading, evaluation protocols, tuning regimes, and support for systems with custom tuning, inspired by established tabular benchmarks such as TabArena. We plan to work with the research community to further develop RelArena-α into a catalyst for progress in the relational learning community. We release the initial version of TabPFN-Rel, a purpose-built relational harness for TabPFN-3. Currently ranked first among models on RelArena-α, TabPFN-Rel makes key improvements upon RDBLearn. Beyond its ranking, TabPFN-Rel serves as a strong baseline, adding to the growing evidence that flattening a relational database into a single table remains competitive with specialized relational architectures on real-world tasks. To facilitate adoption of relational learning methods in research and industry, we release an initial α-version of our Relational Predictive Interface, RPI, an open-source, model-agnostic interface that enables early adopters to easily define problems on new databases and apply any model implemented in RelArena-α, including TabPFN-Rel, to these problems.

AI has long assisted scientific research, but the rapid advance of LLMs and agentic scaffolds is reshaping the landscape; a single system can now carry whole-stage research from an initial hypothesis all the way to final published paper, which is a paradigm now referred to as AutoResearch. Existing evaluations reveal little about how these agents operate or where they break down. Tasks are narrowly-scoped, evaluation measures performance but not process, and failure diagnoses lack systematic coverage or artifact-level visibility. To address this gap, we introduce AutoResearchEval, featuring 100 tasks grounded in published frontier science across 7 scientific domains and the full research lifecycle, including ideation, retrieval, execution, analysis, writing, and review. Evaluating 8 harness-model combinations yields 800 autoresearch agent trajectories, with process-level annotation. We organize these insights into AutoResearch Failure Taxonomy or ARFT, a framework of 45 empirically-grounded failure patterns. To enable scalable fine-grained attribution, we leverage a human-calibrated agent-as-a-judge pipeline to inspect complete trajectories and intermediate artifacts. Failure patterns converge on a single overarching limitation, namely that current agents lack a metacognitive loop, which entails the ability to check what they produced against what they found, revise when it does not hold up, and question whether the path they took was sound. The same patterns recur across all 8 harness-model combinations, including the strongest models tested, locating the deficit at the model level rather than in any particular scaffold; whether orchestration-level interventions can close it is an open question this work does not test. We publicly release AutoResearchEval and ARFT to facilitate continued research and development in autonomous scientific discovery.

Large language model (LLM) agents are evolving from conversational assistants into autonomous systems that execute long-horizon tasks through reasoning, tool use, code generation, and workspace manipulation. As agents increasingly operate over persistent environments and multi-step workflows, they face challenges analogous to those addressed by transactional database systems: reliable execution, consistent outcomes, safe concurrency, and durable state management. We introduce the concept of an agentic transaction and propose an ACID-compliant agent system framework that reinterprets the classical ACID properties for agent execution through four semantic guarantees: Semantic Atomicity, Semantic Consistency, Semantic Isolation, and Semantic Durability. Together, these properties provide a principled foundation for building reliable agent systems despite model uncertainty and dynamic execution environments. To instantiate this framework, we develop an ACID-compliant data agent that realizes these guarantees through transactional exploration-execution-validation cycles, transactional skill hubs, confidence divergence-based validation, semantic dependency-aware isolation, and transaction-aware semantic state management. Experimental results on widely used benchmarks show that our system achieves a 10.6% improvement over state-of-the-art agents, including Claude Code. This work opens a broader research agenda on extending transactional principles and system architectures toward building trustworthy, scalable, and self-evolving AI agent systems.

The rapid evolution of text-to-image (T2I) generation models has effectively solved the foundational challenge of raw pixel synthesis, shifting the community's focus toward fulfilling increasingly intricate user requests. While recent agentic image generation workflows enhance static inference with advanced capabilities like external knowledge retrieval and iterative reasoning, they mostly operate in isolated silos with fixed ``one-size-fits-all" topologies. This inevitably leads to severe compute-mismatch, where simple queries are forced through computationally heavy pipelines. To bridge this gap, we present GenRouter, the first unified workflow routing framework for agentic image generation. We first formulate GenCanvas, standardizing diverse agentic pipelines into a universal set of foundational primitives and executable templates. Operating over this unified space, GenRouter adaptively routes heterogeneous prompts to their optimal workflows via (i) demand profiling, (ii) experience matching, and (iii) Pareto filtering. Extensive experiments across diverse benchmarks demonstrate that GenRouter achieves superior visual alignment while reducing execution costs by over 95% and latency by 65% compared to heavyweight static pipelines. Furthermore, the system continuously self-evolves via accumulated experience, enabling robust zero-shot generalization that boosts performance and halves computational overhead.

Frontier agentic systems powered by large language models (LLMs) exhibit human-like patterns of cognition. As these systems become deeply integrated across different domains, their cognitive engagement raises critical concerns for human society that remain insufficiently studied. To address this gap, we systematically analyze risks induced by expanding cognitive capabilities, following a three-level framework defined by their cognitive scope, from physical cognition to social cognition, and finally to self-referential cognition. We study their potential risks to human agency, autonomy, and control capability, corresponding to each cognitive level. We finally propose strategies to mitigate these risks and enhance the controllability of agentic AI systems, ensuring their long-term safe development.

Part-aware 3D object generation is essential for graphics applications such as controllable modeling, editing, and articulation, where objects are represented as coherent assemblies of semantic parts. However, existing part-aware generation methods, do not scale well to highly complex objects. As the number of parts increases, generating detailed geometry becomes prohibitively expensive in token length and memory. We introduce MegaParts, a scalable autoregressive 3D generation framework to address this challenge by combining structured sequence modeling with a token-efficient vector-quantized shape tokenizer. Our tokenizer learns discrete latent representations for part-level geometry by minimizing token usage subject to high-fidelity reconstruction, enabling adaptive-length tokenization based on geometric complexity. On top of this compact representation, we train a large language model to generate object bounding boxes, part bounding boxes, and part shape tokens within a unified structured sequence. Combined with efficient long-context training strategy, our token-efficient formulation scales to objects with up to 300 parts and sequence lengths up to 256k tokens. This substantially extends the scale of part-aware 3D generation while preserving compositional structure and enabling fine-grained part-level control. Our method achieves higher mesh quality than baseline autoregressive and diffusion models, showing that compressed discrete part tokens improve not only scalability but also the achievable fidelity of generated geometry. These results suggest that LLM native token-efficient autoregressive modeling is a compelling alternative to diffusion for large-scale part-aware 3D generation. The project page is available at https://expmaster.github.io/megaparts_webpage.

New York Times : Internal memo: ICE bars its employees from wearing Meta's AI glasses, saying they “could unintentionally capture, record, or transmit sensitive information” —  The agency joins a growing number of workplaces and groups to ban Meta's devices, which have spurred privacy concerns.

Allan Smith / NBC News : Pennsylvania Governor Josh Shapiro signs an executive order imposing new requirements on data center projects, including getting approval from local officials —  Shapiro, a potential Democratic 2028 contender, earlier welcomed data center developments.  His new executive order takes a starkly different tone toward the industry.

Reuters : In opening arguments, US state AGs said Meta intentionally sought to addict children to Facebook and Instagram in pursuit of profit; Meta rejected the claims —  Meta Platforms (META.O) rejected accusations by U.S. states that it intentionally sought to addict children to its Facebook …

Alex Heath / Time : Sam Altman says OpenAI's decision to pace its AI development was caused by a collection of research observations showing “various degrees of misalignment” —  “I think it is a good time to slow down,” OpenAI CEO Sam Altman told me last week, describing the company's decision …

The Information : Sources: Anthropic has been preparing to give its co-founders shares with extra voting power to help insulate them from outside pressure, ahead of a planned IPO —  Anthropic has been preparing to give CEO Dario Amodei and other co-founders a class of stock with extra voting power to help insulate …

Bloomberg : The US SEC proposes exempting certain digital asset offerings from securities registration statements, with exemptions for offerings up to $5M and $75M —  The Securities and Exchange Commission unveiled a proposal to exempt certain digital asset offerings from securities registration statements …

Mary Ann Azevedo / Crunchbase News : VC funding in physical AI companies totaled $47.4B across 521 deals in H1, up 80% from H1 2025, and more than the $41.9B invested from 2022 to 2024 —  Funding to physical AI companies is booming in 2026.  —  Venture investors appear to increasingly see physical AI as the next leg of the broader AI boom.

Ina Fried / Axios : OpenAI changed safety practices and paused RL training for two weeks after the Hugging Face breach and evidence Astra may have met a critical cyber threshold —  OpenAI said Tuesday that it has made several changes to its safety practices following its determination that an upcoming system …

Jennifer Pattison Tuohy / The Verge : Comcast is rolling out Wi-Fi motion sensing to its Xfinity XB7 and newer gateways, a free opt-in feature that enables its routers to detect movement at home —  Comcast is activating free Wi-Fi motion sensing on Xfinity gateways, allowing its routers to detect movement at home.

Bloomberg : Sources: Anthropic's revolving credit facility is set to surpass its ~$10B target as banks compete for roles on the upcoming IPO; Anthropic could limit the size —  Anthropic PBC's revolving credit facility is set to rise above its roughly $10 billion target, according to people familiar with the matter …

Bloomberg : The EU says it “welcomes Apple's changes”; Apple will reduce its fees in the EU, charging 26% for apps with IAP, 20% for alt payments, and 15% for link-outs —  Apple Inc. has agreed to a set of sweeping changes for its App Store in Europe, including dropping per-install fees …

Apple : Apple unveils new EU business terms from October 1, resolving disputes with the EU, including a new 5% Core Technology Commission and App Store commission tiers —  Apple today announced changes to its business terms for apps in the European Union, following close collaboration with the European Commission.

Melia Robinson / Business Insider : Harvey announces Harvey Tenet, its first in-house, proprietary model for legal work, trained on mock disputes and case files using its own version of Kimi K3 —  Harvey built an $11 billion legal-software business on top of other companies' AI models.  Now it's trying to prove it can build one of its own.

Brock E.W. Turner / Axios : Francisco Partners plans to take patient engagement software company Weave Communications private for ~$650M; Weave went public at a ~$1.5B valuation in 2021 —  Francisco Partners plans to take patient engagement software company Weave Communications private for approximately $650 million, the companies announced Tuesday.

Wired : Meta ran ads for a nudify app; one ad showed a porn video with a deepfake closely resembling a prominent US politician; Apple removed the app from the App Store —  One advertisement featured a pornographic video with a deepfake closely resembling a prominent US politician.

Mozilla 释出了 Firefox 154。主要新特性包括：Local Network Access 保护支持 WebSocket 连接，当网站尝试与本地网络的一个设备建立 WebSocket 连接，它们需要先获得许可；Smart Window 能建议相关标签页分组，并为每个分组建议一个名称；Windows、Linux 和 macOS 都支持本地配置文件备份，能在所有三个平台上恢复备份；网站可设置在浏览器关闭时豁免清除其 Cookie 和网站数据；地址栏新增 Manage AI 快捷操作；修复 bug，等等。

Google 以 1000 万美元拍下了破产航空公司 Spirit 的数据。Spirit 成立于 1980 年，主要经营加勒比海地区、拉丁美洲等定期国际航线。该公司在新冠疫情后遭遇财政困难，2026 年 5 月宣布倒闭，公司进入清算程序，目前正拍卖资产以筹集资金和偿还部分债务。法庭文件显示，Google 以 1000 万美元拍下 Spirit 的数据，包括 1 亿封电子邮件、5 亿条 Microsoft Teams 信息、1700 万份 OneDrive 文件，2050 万 SharePoint 文件，逾 3000 万则客户服务电话录音和逾 1500 万则客户服务聊天记录，60 万个 ServiceNow 工单，以及逾 76.3 万次航班、500 万组机组人员、逾 120 万张燃油单据，787452 个零部件采购记录等公司运营数据，等等。Google 表示购买这些数据是为了改进其 AI 服务。法庭文件显示，这些数据在出售前已进行匿名化处理，Google 也承诺会清除其中发现的任何个人身份信息。

根据在 macOS Tahoe 26.7 RC 版本发现的一则视频，苹果正在研发配备摄像头的耳机 AirPods，该产品处于即将发布状态。AirPods 的摄像头可以在 AI 帮助进行目标识别，在演示视频中，苹果通过摄像头识别一本书的书名。摄像头会将信息传输给 Visual Intelligence‌，Siri 将能回答佩戴者周围环境的问题并记录信息。配备摄像头的 AirPods 代号为 B790，最早可能会在 9 月宣布。

根据 FDA 上月发布的年度调查报告《2025 National Youth Tobacco Survey》，2025 年美国青少年吸烟率降至历史新低的 1.4%。该年度报告自 1999 年以来一直在追踪 11-18 岁美国青少年吸烟情况。美国青少年吸烟率历史最高水平发生在 1997 年，当时有 36% 的高中生表示正在吸烟，70% 表示至少尝试过一次吸烟。美国青少年的电子烟使用率也在 2025 年降至 5.2%。美国成年人的吸烟率保持相对稳定。老年人的吸烟率仍然高于年轻人，他们面临的吸烟相关风险最大。吸烟率的下降可能主要归功于公共卫生政策和关于吸烟危害的公共宣传。但最近 Make America Healthy Again 运动将尼古丁宣传为一种健康产品，未来吸烟率是否回升有待观察。

老年人在极高温度下更易面临危险。因为他们出汗较少，心脏在高温下需要更努力工作，且血管扩张能力较弱，限制了保持体温凉爽的能力。随着全球气温上升，以及预计到 2050 年全球 60 岁及以上人口将翻一番达到 21 亿，这意味着会有更多老年人将遭受高温相关的健康问题。发表在《The Lancet Public Health》期刊上的一项研究预测老年人将比年轻人面临更频繁、更广泛的不可补偿热应激（uncompensable heat stress）。研究表明，气温仅比工业化前水平升高 1.5C 老年人就会面临有害甚至致命的热应激风险，而年轻人需要升高 4C 才会面临相同风险。全球气温每升高 1C，每年就会有约一百万人暴露在至少 180 小时的极端高温环境中。老年人承受的负担尤为沉重。研究人员表示，如果全球气温持续升高，热带和亚热带部分地区可能变得不适宜居住，尤其对老年人而言。

首都医科大学附属北京中医院的研究人员报告了一个极其罕见的病例：一名 40 岁的女子同时患有四种自身免疫性疾病。她此前已诊断有溃疡性结肠炎——会导致大肠内壁发炎，可能引起疼痛、腹泻和出血；自身免疫性肝炎——自身免疫系统错误攻击并引发肝脏炎症的疾病，与病毒性肝炎不同，它并非由感染引起；原发性硬化性胆管炎。她因为严重缺铁性贫血就医，医生在仔细检查后确认她患有第四种自身免疫性疾病——自身免疫性胃炎，缺铁性贫血源于自身免疫性胃炎导致的铁吸收障碍。医生给她开了铁剂。

Mass General Brigham 医院的一项研究发现，每天至少饮用一杯含糖饮料会使胃癌风险增加两倍以上，很少饮用含糖饮料的人风险较低。研究分析了 112,284 名参与者的数据，收集了参与者数十年的饮食、生活方式和健康信息。在长达数十年的随访期间，共有 278 名参与者罹患胃癌。在考虑了其他潜在风险因素后，相比每月饮用不到一份含糖饮料的人，每天至少饮用一份含糖饮料的人患胃癌的风险高 2.45 倍。这种关联性在男性和女性中均观察到。较高的总果糖摄入量也与更高的胃癌发病率相关，果糖是含糖饮料使用的主要甜味剂。胃癌是全球第五大癌症死因。

美国 30 个州联合起诉社媒巨头 Meta 的案件将于 8 月 18 日开审，若 Meta 败诉，已成为社交媒体日常使用核心的功能都可能被迫改变，公司还可能面对高达 1 万亿美元的索赔。这起诉讼由包括加利福尼亚州和纽约州在内的 30 个州于 2023 年提出，涉及多项联邦和州儿童隐私法规。各州认为，Meta 明知旗下平台可能对年轻用户造成伤害，却仍利用相关功能让儿童和青少年长时间停留在平台，以扩大用户群和商业收益。诉方要求 Meta 对未成年用户使用 Instagram 和 Facebook 的方式作出一系列改变，包括取消点赞数和无限滚动、停止视频自动播放，并调整被指会刺激用户持续使用平台的推荐算法。各州还要求 Meta 建立家长验证机制、移除部分会改变用户外貌的图片滤镜、禁止未成年用户开设多个账号，以及取消 Instagram Stories 等阅后消失内容。诉方认为，这些如今已成为社媒日常体验的功能，很多本身就是为了提高使用频率和延长用户停留时间。Meta 否认相关指控。

Anthropic CEO Dario Amodei 表示，对 AI 日益高涨的抵触情绪根源在于企业、政府和科技行业普遍存在的信任危机。Amodei 承认公众对 AI 持负面看法，认同这是一个大问题。但他不同意这种负面看法主要是由他或其他任何 AI 行业领袖对 AI 风险的警告造成的，“我认为这从根本上来说是一场信任危机。我认为普通民众不信任公司、政府或科技行业，总是怀疑我们在谋划某种新方法去坑他们。”

法国独立开发者 Théo Ducreux 创建的 ValidateHTML 项目检查了最广泛使用的 5000 个 Web 域名的内容，这些域名既有 Google、YouTube 和 Akamai 等巨头，也有不面向公众的如 Google 的 gstatic 和 EZVIZ 的 Ezviz7。Ducreux 发现，2656 个网站有人类可读的主页，还有很多域名则只是流量重定向网站。近九成的网站未提供符合 W3C 和 WHATWG (Web Hypertext Application Technology Working Group)规范的 HTML。统计显示，所有网站包含 100,305 处违反 HTML 规范的错误，有 18,863 处 CSS 错误。只有 12.8% 的网站有完全有效的 HTML，只有 2.6% 的网站完全符合规范。逾三分之一的网站未通过无障碍检查。

2021 年 12 月 Better.com CEO Vishal Garg 通过 Zoom Call 解雇了 900 名雇员，约占员工总数的 9%。Garg 在 Zoom Call 中告诉员工，你是不幸被裁员的成员之一，你的工作被终止，即时生效。Better.com 是一家抵押放款公司。五年之后，Garg 也成为了被解雇的对象，而且他还想要恢复原职。Garg 于 8 月 3 日被董事会解雇，接替他的 Daniel Lewi 是在一周前在他帮助下进入董事会的。在其领导下，公司市值从 80 亿美元暴跌至 3 亿美元。尽管如此他的律师致函董事会，要求恢复 CEO 职位。信中称Garg 将通过一项新计划扭转公司颓势，该计划旨在训练 AI 处理抵押贷款业务。

亚马逊正在大量采购图书，拆开书脊扫描书页，在此过程中销毁图书。404 Media 在一本怀疑会被 AI 公司收购用作训练数据的珍本图书内放置了跟踪设备 Apple AirTag，一路跟踪到了目的地。最终目的地是位于内华达州拉斯维加斯的一个亚马逊仓库。在该仓库工作的亚马逊员工表示，他们的工作是接收大量印刷版图书，拆开书脊快速扫描图书。印刷图书在扫描过程中会被销毁。该仓库工作的亚马逊团队被称为 VGT3，其 logo 是一只张着嘴、手里拿着一本书的恐龙。

根据 Hugging Face 的报告，阿里巴巴旗下的开放权重模型通义千问（Qwen）过去半年全球下载量突破 30 亿次，高居第一。相比下 Google 的开放权重模型下载量为 4.18 亿次，Meta 为 2.27 亿次。Hugging Face 称，2026 年几乎每个月中国 AI 实验室都会发布高性能开放权重模型，模型参数上限 7540 亿到 2.78 万亿之间，相比下美国实验室的开放权重模型参数通常低于 1300 亿，少数例外是英伟达发布了 5610 亿参数的 Nemotron 3 Ultra 模型，Thinking Machines 发布了 9520 亿参数的 Inkling——该模型是基于中国模型构建。中国发布的 178 个 200 亿以上参数的模型中，55% 使用了 Apache 2.0 许可证，22% 使用了 MIT 许可证，大部分有非商业使用限制。最近发布的 Kimi K3 和 Qwen3.8 大型模型除了有非商业使用限制外，还有收入分成要求。报告将阿里巴巴的 Qwen 称为是开放 AI 生态系统最大的社区基础模型，有 151,448 个衍生模型是基于 Qwen，其次是 Google 有 82,506 个衍生模型。

OpenAI 举报了一名在与 ChatGPT 聊天中透露奸杀前女友计划的佛罗里达男子。Darren Zhou 告诉 AI 机器人，“我打算在这个月底之前杀了她，如果我得不到她，谁也别想得到她。”Zhou 是一名被高盛解雇的分析师，他在 3 月开始与 ChatGPT 谈论分手的女友，描述了女友的爱好、常去的地方、对女友身边其他男人的嫉妒以及想复合的愿望。OpenAI 的政策规定，当公司检测到用户计划伤害他人时，一个小型团队会审查互动记录。严重的情况下可能会举报给执法部门。Zhou 的聊天提及了强奸、谋杀和谋杀后自杀。联邦探员于 5 月将其两个月的聊天记录转交给当地警方。警方检查了 Zhou 前女友的安全状况。他的前女友 21 岁，因 Zhou“反复无常、嫉妒心强且控制欲强”而与他分手，因为害怕而通过电话分手。警方于 5 月将其逮捕。他在县监狱待了两天后缴纳 10 万美元保释金后获释。他与检方达成认罪协议，于 8 月 13 日对三项指控认罪，被判处 8 年缓刑，缓刑期间须遵守一系列限制。

中文维基是最大的非官方中文百科，它通过公开讨论决定条目的存废，编辑们引用政策进行辩论，最终由一名编辑审阅整个讨论并对条目最终命运做出具有约束力的决定。但究竟是谁参与了讨论？决策权的集中程度如何？哪些议题会引发争议？又是什么逻辑驱动最终结果？纽约城市大学的两名研究人员分析了 2020-2024 年间中文维基 47,030 个条目存废讨论案例，发现虽然每年有数百名编辑参与讨论，但最终决定权集中在少数人手中，而且这些决定者的人数还在不断减少，其中一人在一年内做出了 89.9% 的决定。研究还发现，虽然简体中文用户人数是繁体中文人数的 45 倍，但中文维基中繁体中文编辑人数是简体中文编辑的 1.2-1.8 倍。删除条目的最主要理由是缺乏知名度，逾四分之一删除没有给出明确的理由。研究人员发现 6 名中文维基管理员做出了 85%- 100% 的决定。2023 年 Shizhao 处理了 6,312 个案件，每天 17.3 个，占到了总决定的 89.9%；2024 年 Shizhao 和 Manchiu 总共处理了 86.5% 的案件；三位管理员蟲蟲飛、Lanwi1 和 AT 在 2021-2023 年离职，而 Manchiu 在 2022-2023 年间处于非活跃状态，2024 年回归。当编辑们投票保留条目时 Shizhao 倾向于删除，而编辑们投票删除条目时 Manchiu 倾向于保留。两人处理了 86.5% 的决策，他们的个人倾向会影响最终结果的分布。