OrangeBot.AI Digest — 2026-08-16

Wake at the end of your sleep cycle

Open-source unified interface for agent harnesses

Healthspan score from your wearables, labs, and DNA.

AI-ready, Markdown-first documentation framework

AI-powered construction takeoffs, estimates, and invoices

The Interactive 3D Encyclopedia

Vapi for FaceTime: AI video agents in a few lines

Close your MacBook. Keep the work running.

A chief of staff in your pocket

Your favorite uncle who knows sports & bets, on iMessage

the lo-fi of LLMs: your mind, made larger

Coding leap from scaled post-training on the same base

Bringing intelligence to your cursor

A Mac menu bar app that throws confetti for your wins

Scheduling and calendar sync, all in one

Interactive world models must support persistent memory, responsive interaction, and long-horizon generation, yet these requirements place conflicting demands on the model. Maintaining history in the denoiser context or key-value cache incurs growing cost, forcing a trade-off between session length and retained memory, while low-latency interaction relies on few-step generation whose capabilities are bounded by its teacher. Evoke addresses both limitations by externalizing persistent world state and redesigning the teacher for long-horizon interactive generation. Scene geometry is maintained in an external, camera-indexed world state bank, from which only view-relevant information is retrieved, keeping the denoiser context bounded as the session grows. Rather than treating the teacher as a fixed generator, we design it for long-horizon supervision: its sparse attention combines chunk-wise grouping, retrieval of selected distant frames, and a linear-attention global state, yielding linear growth in memory and compute while enabling supervision over long horizons. Such supervision exposes content drift that stays locally plausible within short windows, while per-chunk conditioning enables prompt changes and event control throughout the sequence. A 30-second distribution-matching objective, applied under self-forced rollouts, transfers both capabilities to a three-step student that uses no classifier-free guidance, improving resistance to long-term drift while preserving responsive conditioning. With bounded context and recurrent external memory, Evoke supports open-ended, continuously evolving generation; on a single H200 at 384times 640, each 1.5,s chunk is generated in 2.11,s. As a three-step world model, Evoke achieves state-of-the-art performance on WBench while remaining competitive on VBench-Long and VBench-2.0.

No single large language model (LLM) is optimal across all queries and budget constraints, making model routing essential for cost-effective deployment. Existing routers adopt diverse formulations and implementations, making fair comparison and extension difficult. We present a unified formulation of LLM routing as a sequential decision process characterized by five components: context encoders, model encoders, scoring functions, decision rules, and learning signals, covering single-turn, multi-turn, and personalized routing. Based on this formulation, we develop an automated pipeline for constructing routing supervision and evaluating routers jointly on response quality and inference cost. The resulting benchmark, xRouteBench, spans generic LLM, memory-augmented, vision, time-series, and personalized routing tasks. We further introduce LLMRouter, an open-source modular infrastructure with more than 16 representative routers. Our empirical study shows that learned routers outperform the strongest fixed-model baseline by 14.6% relatively, lightweight routers become more competitive under tight cost constraints, and user-conditioned routing consistently improves personalization.

We present DreamX-Phi 1.0, an action-conditioned video world model for robotic manipulation that, given an observed frame, a language instruction, and a prescribed action sequence comprising end-effector poses and gripper states, predicts the resulting future observations. Yet realism alone does not guarantee faithfulness: a convincing rollout can still move the wrong arm or lose the manipulated object. To ensure the prediction respects each arm's commanded path, we inject per-arm SE(3) transformations into attention via PRoPE-style geometric encoding, preserving arm identity and rigid-motion structure. Action control alone does not fully constrain scene geometry or the evolution of small manipulated objects. We therefore add a lightweight depth branch for scene-level geometry and use SAM3 masks with a frozen V-JEPA teacher to maintain object consistency throughout grasping. We further distill the multi-step generator into a few-step student via distribution-matching distillation for efficient deployment. At the time of writing, achieves first place on Track~1 and second place on Track~2 of the WorldArena~2.0 Challenge. Our model and code will be publicly available.

An LLM agent's capability depends not only on model weights but on its harness: prompts, tools, skills, and control flow. Self-improvement loops already edit harnesses, yet single-lineage search is path-dependent and local wins often regress other tasks. We introduce DarwinX, which treats self-evolution as selection over a population of harnesses with the model frozen: a preserve-and-extend contract admits only variants that extend coverage without regressing, an archive keeps alternative lineages for recombination, and failure-, teacher-, and self-derived evidence share one edit interface. Fitness comes from each benchmark's own verifier: no gold solutions, no hand-picked winners. Across four benchmarks that progressively separate the evolution signal from the test, one loop adds about 17 points on average: Terminal-Bench 2.1 rises +7.7 to 83.2% on a matched base and to the verified frontier at 84.7% on a stronger one; TerminalWorld's held-out split reaches 68.3%, ahead of every off-the-shelf agent; WebArena-Infinity real-task pass@1 rises from 43.5% to 93.0% audit-clean; and a Terminal-Bench 2.1 harness transfers unchanged to SWE-bench Verified. What evolves is general agent competence, not benchmark-specific patches, so it survives changes of task, verifier, and base model. A frozen model need not be a fixed agent: harness selection turns evaluation compute into durable capability.

Scientific discovery increasingly requires AI systems that can reason over scientific evidence of heterogeneous modalities, interact with scientific tools and environments, and sustain progress across long task horizons. We present Intern-S2-Preview, a series of scientific agentic foundation models designed to support multimodal scientific understanding, reasoning, generation, and long-horizon tasks. The training pipeline begins with scientific multimodal pre-training over rendered scientific documents, interleaved image-text data, and diverse scientific corpora. Starting from the pretrained checkpoint, we apply a unified post-training pipeline consisting of supervised fine-tuning, scalable multi-task reinforcement learning (RL), black- and white-box agentic RL, and on-policy distillation. This pipeline is supported by practical techniques that improve rollout and training stability and efficiency, including partial rollout with off-policy correction, adaptive length regularization, online speculative decoding, robust multi-task optimization, and trace-aware experience assembly for agentic tasks. At the architecture level, Intern-S2-Preview-397B extends time series modelling from efficient long-sequence understanding to numerical forecasting, while Memory Decoder is studied as a separate memory-augmented path for rapid scientific specialization without modifying the frozen 397B backbone. Evaluations across scientific, multimodal, agentic, and general-purpose benchmarks show that Intern-S2-Preview-397B achieves competitive or leading results in multiple settings. The time series modules improve scientific signal understanding and forecasting on SciTS, while the separate Intern-MemDec-4B extension improves the Biology-Instructions average score from 56.92 to 60.32 without modifying the frozen 397B backbone.

As large language models increasingly participate in scientific evaluation, we investigate a potential form of reward hacking: how rhetorical choices shape AI-review judgments when reported scientific content is preserved and how these effects vary across evaluation conditions. We construct a controlled corpus of 4,200 full-paper manuscripts derived from 120 anonymized ICLR 2026 submissions. Two LLM rewriters transform six rhetorical dimensions in opposing directions, and five LLM reviewers evaluate the resulting manuscripts under standard and strict protocols. We also test joint, recursive, and reviewer-guided rewriting. Our results show that rhetorical sensitivity is structured rather than uniform. Evidence framing and novelty stance produce the largest positive-negative contrasts in overall assessment, with scope framing forming a weaker second tier; the remaining dimensions have smaller or less stable effects. This hierarchy persists across human-assessed quality levels, but score movement depends strongly on the AI reviewer's original score: lower scores tend to rise, higher scores tend to fall, and directional contrasts are clearest in the middle ranges. More elaborate workflows do not reliably yield larger gains. Joint rewriting is strongly rewriter-dependent, reviewer guidance does not consistently outperform an unguided second pass, and repeated rewriting yields diminishing, configuration-dependent returns. Across conditions, the rewriter primarily determines the separation between opposing variants, whereas the reviewer determines the magnitude and sign of their score effects. Strict review lowers mean OA by 1.36 points without consistently changing rhetorical sensitivity. These findings identify when rhetorical presentation influences AI scientific review and motivate evaluation systems robust to content-preserving variation in scientific writing.

Transforming multimodal sources into condensed and structured media outputs can be fundamentally conceptualized as a long-horizon agentic process centered on a model-harness system. While an ideal harness system should align with human design priors and accumulate reusable experience through empirical exploration to drive recursive self-improvement, existing paradigms remain static and fall short of this capability. In this paper, we present AutoDesign, a framework that aligns with human design priors, where a meta-harness optimizer guides a code agent to recursively improve harness based on rollout feedback. To instantiate and evaluate this framework, we focus on the academic paper-to-poster generation task and introduce PosterBench, comprising a 100-paper Main Track spanning five disciplines and PosterBench-mini, a shared 10-paper subset for controlled evaluation. On the PosterBench Main Track, AutoDesign achieves the highest score of 78.32, surpassing the closed-source commercial system Claude Design by 7.45 points. Across seven controlled code-agent-model configurations, integrating the learned DesignHarness consistently improves performance, increasing the average PosterBench Score from 54.99 to 67.39 (+12.4%). In a fully autonomous long-horizon loop, it executes 253 tool calls and 11 editing turns within 40 minutes for under $3, reaching average conference-poster quality in human evaluation. A system-blind human study further demonstrates that AutoDesign achieves the highest human preference among evaluated systems.

Video world models simulate future states conditioned on current observations and user actions. Recent systems have demonstrated impressive video consistency and action controllability over long sequences. However, fairly comparing these interactive models remains challenging. In practice, a human player typically evaluates a world model by pursuing long-horizon objectives through interaction. For example, a user may turn around 360 degrees to see whether the environment remains consistent, or walk into the water and inspect whether realistic water ripples are generated. The action sequence required to achieve the same objective may vary substantially between models, making fixed action-conditioned evaluation unsuitable for cross-model comparison. To address this, we employ multi-modal Agent Players to interact with world models toward specified long-horizon objectives. Building on this paradigm, we introduce PlayWorld, a benchmark providing 171 scenarios, each with a specified objective. To evaluate performance thoroughly, we assess models along four core dimensions: geometry consistency, interaction fidelity, out-of-sight evolution, and insight evolution. In addition, we incorporate basic ability metrics for video quality and controllability. Experiments across nine state-of-the-art world models reveal that current models remain unreliable on long-horizon interactive objectives, particularly in maintaining spatial consistency and persistent state evolution. Code and data are available at https://github.com/kxding/PlayWorld.

Spatial intelligence is becoming a foundation for embodied agents, robotic planning, and multimodal assistants. To improve the spatial reasoning ability of VLM agents, existing work has mainly followed two lines. One line uses post-training methods, such as supervised fine-tuning and reinforcement learning. Another line adopts an agentic paradigm in which the model calls external spatial tools, such as depth estimation and 3D reconstruction tools, to gather intermediate spatial evidence. We study a complementary and underexplored route: Can a frozen VLM agent improve its spatial reasoning through parameter-update-free self-evolution, without depending on external expert spatial tools at inference time? We present Spatial Memory Agent (SMA), an experience-grounded runtime framework that converts verified spatial experience into reusable transferable lessons. In a verifiable spatial environment, SMA queries the frozen VLM, obtains a predicted answer and reward, and uses verifier-guided reflection to distill compact transferable lessons from spatial experience. SMA further assigns each lesson a Transfer Reliability Score (TRS), which is initialized uniformly and calibrated from later retrieval outcomes as visit evidence of future transfer reliability. During read-only deployment, SMA retrieves lessons by semantic filter and similarity-TRS combined ranking, allowing the retrieved memory to guide frozen model inference. Across five representative spatial benchmarks and four base VLMs, SMA achieves the highest macro average in every base-model block and the best accuracy among the evaluated methods in most of the 20 evaluations, establishing a practical parameter-update-free path for spatial self-evolution across the evaluated frozen model scales and environments.

We present the first systematic study of Massive activations (MAs) in layer-interleaved HLA LLMs and uncover two architecture-aligned morphologies: MAs consistently spike immediately before full attention layers, forming pre-attention spikes (PAS), and can persist through intervening linear attention layers, giving rise to inter-spike plateaus (ISP). As full attention becomes denser, successive PAS become increasingly connected through ISP, ultimately recovering the stable MA morphology of full attention LLMs. We establish the recurrence of this organization across five linear attention architectures, six hybridization configurations, five data domains, and representative open-source hybrid models spanning 1.2B to 397B total parameters. Controlled pretraining of GDN-based hybrids at scales up to 1.3B shows that both morphologies emerge early and respond asymmetrically to output gating: full attention output gating strongly attenuates their absolute magnitudes without eliminating their layerwise organization, whereas removing GDN gates yields comparatively modest amplification. Mechanistically, our systematic-outlier analysis supports a shared lifecycle account governed by the timing of MA cancellation. PAS follows a localized write-sink-cancel process, while the extended persistence of ISP is consistent with delayed cancellation. At the full attention limit, this account recovers the stable MA morphology characteristic of full attention LLMs. Our code is available at https://github.com/StartluxLabs/Massive-Activations-HLA.

Pose-driven human animation synthesizes a video of a target person from a single reference image and a driving pose stream. Real-time generation is essential for interactive applications such as live streaming, telepresence, and virtual avatars, yet diffusion-based systems require minutes to hours per clip, precluding responsive interaction. We present LiveAnimate, to our knowledge the first animation system to combine real-time streaming with stable long-form generation at billion scale, built on a 14B-parameter video Diffusion Transformer (DiT). A two-stage training pipeline first adapts a pretrained bidirectional DiT into a block-causal autoregressive generator through Reference-Anchored Teacher-Forcing Adaptation, and then reduces the sampling budget to three steps through Block-wise Self-Forcing Distillation. To preserve appearance over extended streams, we introduce Pose-Retrieval Sink Attention (PR-Sink), a bounded KV-cache mechanism combining a Static Sink that permanently anchors the first generated block, a Dynamic Sink that holds a pose-retrieved historical block, and a three-slot Rolling Window. When a pose recurs, PR-Sink restores the relevant appearance context without retaining the entire sequence, so memory and per-block latency remain constant regardless of stream duration. Together with Ulysses sequence parallelism and operator fusion, these designs enable 19.63\,FPS streaming inference on two NVIDIA H100 GPUs. On a three-minute benchmark, LiveAnimate maintains nearly constant perceptual quality and identity from the first 30 seconds to the final minute, while prior systems degrade substantially or require hours of offline computation for the same rollout. These results establish a new operating point in quality, latency, and duration for interactive full-body animation.

Talking-video character replacement requires coordinated transfer of appearance and voice while preserving the source motion, scene, linguistic content, and audio-video timing. Existing methods use separately optimized models for the two modalities, making audio-visual consistency difficult to enforce. We present UniSwap, the first framework for streaming joint audio-visual identity replacement in talking videos. Given a source video, a reference image, and a reference voice clip, UniSwap transfers the reference appearance and vocal timbre within a single audio-visual diffusion transformer while preserving the source content and dynamics. To address the scarcity of aligned cross-identity training pairs, we introduce a swap-and-reconstruct pipeline that removes visual and vocal identity from real clips and uses the original clips as reconstruction targets. Starting from a bidirectional backbone, we progressively adapt the model through In-context Pretraining for joint replacement, Conditional Streaming Adaptation for block-causal KV-cached generation, and Efficient Self-forcing DMD for mitigating exposure bias and reducing sampling from 30 to 3 denoising steps per block. Efficient Multi-LoRA Switching enables the three DMD roles to share a single frozen backbone. Feature-RoPE Decomposition keeps cached positions within the training range, supporting stable long-form inference. Experiments demonstrate strong audio-visual synchronization, competitive identity preservation, efficient streaming, and stable long-form generation.

Autoregressive transformers compute along two axes: horizontally across generated tokens, and vertically through model depth. Dense attention gives each token broad horizontal access to the past, but the vertical feedback channel between decoding steps remains narrow: only the sampled token returns to the bottom of the stack, while the top-layer hidden state is discarded. We introduce the full-bandwidth transformer, which widens this channel with latent feedback: at each decoding step, the previous top-layer hidden state is fused with the sampled token embedding through a gated linear unit and fed back as the next input. Latent feedback lets non-verbalized computation re-enter the stack with a renewed depth budget, while preserving the standard transformer architecture, KV cache, and language-modeling objective. To train full-bandwidth transformers without losing parallel teacher forcing, we use a scheduled multi-pass objective that introduces latent feedback late in pretraining and mixes a small fraction of deeper feedback passes for stability. We train 1B-parameter full-bandwidth transformers up to 400B tokens and find that latent feedback improves validation loss, 5-shot language-model evaluation, math and coding generation, and instruction-tuned performance. With negligible per-token decoding overhead, full-bandwidth transformers match or approach standard transformers trained with roughly 1.5times more tokens, and manage to produce shorter reasoning traces at equal or better accuracy.

Visual-token pruning can substantially reduce the inference cost of multimodal large language models (MLLMs), yet existing methods largely rely on fixed, handcrafted heuristics and costly expert trial and error. As pruning objectives, budgets, and model architectures diversify, manually navigating the expanding design space becomes increasingly difficult. This paper aims to build an AI4AI framework for visual-token pruning by addressing a natural question: Can large language models automatically design effective visual-token reduction algorithms? Although LLMs possess broad algorithmic knowledge and strong reasoning capabilities, translating such general knowledge into effective solutions for a specialized task remains nontrivial. We argue that the key lies in designing an appropriate search-state representation that connects the internal knowledge of LLMs with the structural requirements and constraints of visual-token pruning. Based on this insight, we propose AutoPrune, a training-free framework for LLM-driven visual-token pruning policy design. At its core, AutoPrune introduces a Token Pruning Domain-Specific Language (TPDSL) comprising 131 reusable atoms for budget control, token scoring, selection constraints, and token reassembly. A key property of TPDSL is that it represents each search state as a residual modification of a strong base policy. This residual formulation narrows the search space and directs the LLM's attention toward the policy components that are most consequential for performance. Experiments on 14 multimodal benchmarks and three MLLM backbones demonstrate the effectiveness, efficiency, and transferability of AutoPrune. Even when removing 94.4% of visual tokens, AutoPrune preserves more than 99% of full-token performance while reducing FLOPs by 9.9x and prefill latency by 6.4x.

Large-scale manipulation data is essential for robot learning, yet collecting robot demonstrations remains expensive and difficult to scale. Meanwhile, abundant egocentric human manipulation videos provide rich behavioral experiences, but transferring them across embodiments remains challenging due to differences between human hands and robotic end-effectors. Recent advances in video world models offer a promising pathway to synthesize robot-centric manipulation videos from human observations, while their cross-embodiment transfer capability remains largely unexplored. Therefore, we introduce H2R-Bench, a benchmark for evaluating cross-embodiment human-to-robot manipulation video generation, where models transform egocentric human demonstrations into robot manipulation videos under specified embodiments. Each benchmark instance contains a human demonstration video, target embodiment constraints, and source-grounded annotations covering task goals, action events, functional contacts, and object responses. H2R-Bench evaluates generated videos through five dimensions, including goal-state completion, action-event completion, functional contact transfer, embodiment correctness, and general video quality. We benchmark eleven state-of-the-art video generation models across six manipulation families and two robot embodiments. Our evaluation reveals that current video world models remain limited in human-to-robot manipulation transfer: even leading models often fail in embodiment consistency, functional interaction, and task execution. H2R-Bench provides a systematic diagnostic framework for evaluating whether video world models can bridge the human-to-robot embodiment gap and convert human manipulation observations into robot-centric training resources.

Phil Serafino / Bloomberg : French PM Lecornu will hold a crisis meeting Monday over a June-July cyberattack on France's tax agency that compromised 678K individual and business accounts —  French Prime Minister Sebastien Lecornu will hold a crisis meeting Monday to respond to the cyberattack disclosed last week targeting the country's tax collection agency.

Bloomberg : Sources: Stripe has finalized a deal to acquire AI model marketplace OpenRouter for more than $7B; OpenRouter was valued at $1.3B in May —  Stripe Inc. has finalized an agreement to acquire OpenRouter Inc., a startup that helps companies switch between artificial intelligence models …

Alex Janin / Wall Street Journal : Patients, families, doctors, and nurses are increasingly turning to AI tools, such as Face2Gene, to help identify rare and hard-to-diagnose diseases —  Patients, doctors and nurses are turning to the technology for help identifying rare and undiagnosed diseases

Washington Post : How concerns about AI became an important midterms issue, with candidates adding AI and data center policies to their websites in ~40% of races across the US —  Across the United States, political candidates are grappling with an issue that has never before featured widely in American political life: artificial intelligence.

Jason Douglas / Wall Street Journal : A profile of Sony CEO Hiroki Totoki, who aims to transform the company into a business focused on music, movies, video games, and the tech that underpins them —  The company built its empire on electronics.  CEO Hiroki Totoki says its future is entertainment

Owen Walker / Financial Times : Malaysia's 6% Q2 GDP growth was powered by 7.5% manufacturing growth, driven by chipmaking, and 6.6% construction growth, supported by data center development —  Southeast Asian country emerges as key AI hub in boost for its economy  —  The AI boom has made global superstars out of a handful …

Timothy Gowers / Gowers's Weblog : Fields Medalist Timothy Gowers says most famous mathematics problems solved by LLMs so far have almost all been with counterexamples rather than proofs —  For the sake of anyone who might read this blog post in the distant future (a month from now, say), let me mention that I am writing …

Hugging Face : Hugging Face says developers made 151K+ derivatives based on Qwen models, topping others, making Qwen one of the largest foundations in the open model ecosystem —  In the AI world, time feels compressed.  A few months after our spring report in our biannual analysis worked through the ecosystem …

Ilya Gridneff / Financial Times : BlackBerry posted a positive cash position in Q1, its first cash-positive first quarter in any fiscal year since 2017, driven by its QNX and Secusmart products —  A decade after the disappearance of its handsets, the company's future rests on car and security software

Antoine Tardif / Unite.AI : Pathway, which is developing AI models based on what it calls its “Post-Transformer” BDH architecture, raised a $30M seed at a $500M valuation —  AI research company Pathway has secured additional funding at a $500 million valuation, bringing its total seed financing to $30 million …

Grace Shao / Bloomberg : Chinese citizens are more optimistic about AI than Americans because AI is seen as a practical tool in China that disrupts a smaller share of its population —  Different experiences with technology have shaped radically different expectations of what artificial intelligence will bring.

Dario Amodei / @darioamodei : Dario Amodei defends his policy proposals, warns open weights won't decentralize power, endorses pre-launch vetting, says real accomplishments will earn trust —  1/2 Thanks Gavin for an especially thoughtful exchange. I don't usually spend much time on social media but I wanted to engage here because it really brings out the heart of an important conversation. First, on regulation, I think that “either concentrate it in the hands of a

Anna Liss-Roy / Washington Post : How Congressional lawmakers and aides are using AI tools with little oversight to write speeches and news releases, sort constituent mail, and more —  The amendment was one of hundreds filed this summer to an annual defense bill making its way through Congress.

Zeyi Yang / Wired : A look at Unitree's G1 and R1, the humanoid robots behind viral influencer accounts worldwide, as Unitree shipped 5,500+ units in 2025 and readies its China IPO —  The Unitree G1 has found online fame as a relatively affordable robot that can charm a crowd.  But can it ever hold down a real job?

The Information : Sources: Nvidia in talks to invest up to $3B in SoftBank-backed data center developer SB Energy that aims to IPO soon and is behind a huge OpenAI campus in Ohio —  Nvidia is in talks to invest as much as $3 billion in SB Energy, the Softbank Group-backed developer of a massive planned Ohio …

欧盟哥白尼气候监测服务机构证实，6 月和 7 月是西欧有记录以来最热的月份，在此期间估计有 25000 人死于高温相关疾病。热浪导致欧洲各地河流大范围干涸，这一情况甚至从太空中都可以观察到。由于河流水位极低，商船不得不减少多达 80% 的载货量以避免搁浅。这一迫不得已的举措扰乱了供应链，增加了运输成本。河流干涸还导致二战期间几十艘纳粹军舰浮出水面。

发表在 PLOS Medicine 期刊上的一项研究显示，穷人比富人更早出现身体上的老化迹象，其中英国贫困人群出现身体机能下降的迹象比同龄富裕人群早 15 年，加拿大则是早 9 年。这项研究涉及了 8500 多名英国参与者以及 22600 多名加拿大参与者的数据，他们的年龄在 50-85 岁之间，参与了 2012-2023 年间的一项老龄化纵向研究。研究期间参与者多次完成了一系列身体功能测试，包括步速、握力、从椅子上站起的速度和肺功能测量。研究人员比较了研究开始时家庭财富处于最低 25% 的参与者和处于最高 25% 的参与者的身体功能评分。他们发现，在所有身体功能评估中，财富最少的老年人的表现都不如财富最多的老年人。平均而言，英国最贫困的 60 岁人群的步速与最富裕的 75 岁人群的步速相同。75 岁时英国同龄人中最贫困和最富裕人群的功能年龄（functional age）相差 12 岁，而加拿大这一差距为 6 岁。

美国的种族平等过去六十年取得了显著进步，但黑人和白人之间仍然存在巨大的健康差距。有研究估计，1999-2020 年的 21 年间，美国黑人人口的超额死亡数为 163 万，潜在寿命损失 8000 万年。系统性种族主义理论认为，种族不平等的维持部分源于个人偏见，部分源于影响几代人资源、机会和生活结果分配的社会制度。由于这些制度根植于特定地域且难以改变，它们会在资源和机会方面形成持久的本地模式，即使奴隶制之类的政策终止后，这些模式仍会长期存在。加州伯克利的研究人员在 PNAS 上发表论文，他们发现 1860 年美国各县奴隶比例与当代黑人白人死亡率差异显著相关。研究人员估计，1860 年奴隶人口每增加 10%，黑人相对于白人每 10 万人的死亡人数会增加约 22 人。研究结果表明，包括奴隶制在内的一系列种族歧视性政策构建了一个普遍有利于白人、不利于黑人的社会，美国需要更采取强有力的补救性、平等主义政策。

俄罗斯新型 S-71“Monochrome”巡航航空被发现使用了英伟达的 Jetson Orin 计算模块。俄罗斯使用的是 2023 年推出的 Jetson Orin NX 16GB 模块，而英伟达早在 2022 年初就退出了俄罗斯市场，这意味俄罗斯是通过其它渠道获得英伟达芯片的。英伟达表示，Jetson 是消费级产品，军事应用不在其预期设计范围内，但它的销售没有什么限制，因此俄罗斯有很多方法获取该芯片。英伟达表示，“虽然我们无法在产品售出后对其进行追踪，但如果我们确定任何客户违反了美国出口管制，我们将采取适当的行动。”

对 1088 名年龄在 18-34 岁美国人的调查发现，绝大多数受访者表示不信任投资 AI 的大公司高管，包括 Elon Musk、Sam Altman 和 Mark Zuckerberg。45% 的受访者认为 AI 会对其职业生涯产生负面影响，只有 10% 的人认为 AI 会对他们有帮助。年轻人对 AI 公司高层也日益表达不信任，Palantir CEO Alex Karp 的信任度最低，81% 的受访者表示不信任他，紧随其后的是 Palantir 董事长 Peter Thiel（79%）、Alphabet CEO Sundar Pichai（75%）和 Anthropic CEO Dario Amodei（75%）。OpenAI CEO Sam Altman、Meta CEO Mark Zuckerberg、英伟达 CEO 黄仁勋和 SpaceX CEO Elon Musk 的不信任度都在 70% 左右，微软 CEO Satya Nadella 略好一点，不信任度为 65%。60% 的美国人认为数据中心建设必须放缓，只有 15% 的人表示希望加快建设速度。

身陷剽窃丑闻的前剑桥大学教授 Jason Arday 在伦敦死亡，年仅 41 岁。Jason Arday 于 2023 年获剑桥大学任命为教育社会学教授，成为该校历史上最年轻的黑人教授。2026 年初，他被曝出论文抄袭和学术不端，被指抄袭行为明目张胆。他此前做出的大量虚假陈述也引起广泛关注。他声称自己三岁时被诊断患有自闭症、全面发展迟缓、失聪及计算障碍，一直到 11 岁仍没有语言能力。他称自己 18 岁前都不懂阅读和书写，入读大学第一年则被诊断患有阅读障碍。他称自己在 35 日内完成 30 场马拉松，帮助慈善机构筹集巨额资金；声称有人去其在剑桥的办公室威胁他，但剑桥大学未找到任何证据证明发生了此事。他于上周辞去了剑桥大学教授职务，但否认辞职代表着自己承认相关指控。警方表示他的死亡被视为是意外，未发现可疑之处。

法国宪法委员会否决了禁止 15 岁以下儿童使用社媒的法案，裁定该法律过度限制了言论自由，且缺乏足够的年龄验证隐私保护措施。总统马克龙(Emmanuel Macron)要求政府修改法案，目标是在 2027 年春季前推出修订版本。宪法委员会认为，一方面有争议的条款过度侵犯言论和通信自由；另一方面未能提供必要法律保障确保尊重私人生活的权利。法国议员于 7 月批准了该法案，让法国成为继澳大利亚之后欧洲首个禁止 16 岁以下未成年人使用社媒的国家。澳大利亚于去年 12 月实施了全球首个青少年社媒禁令，禁止 16 岁以下未成年人访问包括 Facebook、Snapchat、TikTok 和 YouTube 在内的平台。但在数据表明效果不佳后，澳大利亚议员考虑推行更严厉的处罚措施。欧盟也表示，计划寻求更强有力的保护措施以保护儿童免受有害社媒功能的侵害。

根据发表在《Alzheimer's & Dementia: Diagnosis》上的一项研究，GLP-1 减肥药 Semaglutide 将五年预测糖尿病风险降低 26%。研究人员将 2970 名年龄 65 岁以上且有超重/肥胖和心血管疾病但无糖尿病的成年人随机分配到 Semaglutide 组或安慰剂组。Dementia SomaSignal Test(dSST)对非空腹血清样本的分析显示，Semaglutide 将五年预测糖尿病风险降低 26%，二十年预测风险降低 8.8%。结果表明 GLP-1 减肥药可能具有神经保护作用。

MIT 等的天文学家利用韦伯太空望远镜（JWST）在早期宇宙发现了一个极其明亮的红色斑点。该天体形似一颗巨大的恒星，大小与太阳系相当。但它释放的能量是已知恒星能量的 1000 亿倍。这种能量更接近黑洞可能产生的能量。红色斑点代表了一种全新的天体——天文学家称之为“黑洞恒星”。它是黑洞和恒星的混合体——这种组合此前从未被观测到过。该天体很可能是一团极高密度的气体云，其能量来源并非是核聚变，而是中心的黑洞。天文学家估计黑洞质量为太阳的 10 万倍，他们将该天体命名为 MoM-BH*-1，或者是“一号黑洞恒星（black hole star–one）”。

长鑫科技已超过互联网巨头腾讯控股，成为全球市值最大的中国企业，凸显 AI 热潮正推动投资者对存储晶片股的需求。截至周五 A 股收盘，长鑫科技最新市值约 3.58 万亿元人民币。港股方面，由于市场对腾讯不断增加的 AI 投资担忧加剧，腾讯周五最新市值约 4.0 万亿港元。分析认为，这一市值排名的逆转反映出，投资者继续青睐与 AI 相关的硬件企业，而非传统互联网巨头。总部位于安徽合肥的长鑫科技也是全球第四大 DRAM 供应商。DRAM 广泛应用于从手机到先进 AI 服务器等各种设备。

2026 年 8 月 12 日，欧洲大陆经历了数十年以来首次日全食，白昼变成黑夜，或者“太阳被月亮吞了”？欧洲居民使用了多款智能手机去拍摄罕见的日全食照片，结果发现小米 17 Ultra 智能手机相机拍摄的一张照片极其反常，在这张照片里橙红色的太阳上出现了月亮照片中常看到的山脊、陨石坑和线条，也就是相机将太阳误认为是月亮。小米相机显然是利用了 AI 技术去重建照片中的细节，给太阳添加了根本不存在的月球特征。小米不是唯一这么做的手机厂商，此前三星也曝出过类似的丑闻。

恼人的 Facebook 广告一直是广告屏蔽工具过滤的对象，而 Meta 的工程师也不断迭代广告技术以绕过广告屏蔽工具的过滤。这种猫鼠游戏持续了如此长时间以至于广告屏蔽工具的开发者都受不了，流行广告屏蔽扩展 uBlock Origin 的开发者宣布他们放弃继续反制 Facebook 的广告技术。一位开发者说，“我们将不再支持 Facebook。它是一个令人恶心的反用户网站。它所做的就是监视开源项目的公开活动，然后针对性地投放恶意广告。没错，那些拿着七位数薪水的开发者就是在干这个。”开发者表示，任何有兴趣对抗 Facebook 广告投放的用户都可以提交自己的过滤器补丁。

苹果与阿里巴巴合作为中国市场训练了一个大模型。此前苹果倾向于使用中国合作伙伴的模型为在中国销售的 iPhone  等设备提供生成式 AI 功能。苹果的 AI 功能 Apple Intelligence 预计将在未来几个月通过 iOS 更新在中国推出。苹果推出专为中国市场定制的 AI 模型，使​​其在竞争最激烈的中国市场更好掌控 AI 体验。苹果在中国已被华为等本土竞争对手蚕食市场份额，竞争对手凭借 AI 手机迅速占据领先地位。网信办已经登记了苹果的生成式 AI 服务，为 Apple Intelligence 服务首次登陆国行 iPhone 铺平道路。

YouTube 宣布新内容创作者需要达到更高的门槛才能开始通过广告和订阅获得收入。想要通过 YouTube 平台获利的创作者，需要过去一年内至少获得 8000 小时的有效观看时长，或者在过去 90 天内获得 2000 万次的 Shorts 有效短视频观看量。此前创作者只需要过去一年内有 1000 名订阅者和 4000 小时的有效观看时长，或者在过去 90 天内有 1000 名订阅者和 1000 万次 Shorts 视频观看量。这项变更将于 2 月 1 日生效。此次更新不影响已加入 YouTube 合作伙伴计划的创作者。YouTube 表示，这些调整是为了跟上 YouTube 的发展步伐，目前 YouTube Shorts 的日均观看量逾 2000 亿次，电视观看时长逾 10 亿小时。

《Braid》和《The Witness》作者 Jonathan Blow 在社媒上称，在 Windows 上使用 VLC 启动播放一个 MP3 文件需要 33 秒钟，他因此改用了微软官方的 Microsoft Media Player，批评开源软件问题太多。此事引发了广泛讨论。VLC 官方回应称，核心问题与它无关，是 Windows 11 / Microsoft Defender 的一次更新隔离了 VLC 的插件缓存导致的，重新安装 VLC 或清空缓存可解决该问题。Linux 用户称 VLC 启动播放一个 MP3 文件只需要 1-2 秒。其他 Windows 用户建议在 Microsoft Defender 中排除 vlc.exe 也能解决该问题。