Monthly Digest — 2026-09

Every model with a unified interface. Privacy with proof.

Performance monitoring built for Expo and React Native

The phone that literally has a gimbal built in

Build one customer-facing AI agent and launch it everywhere

Give your AI agent a real phone number in 10 seconds

Now open source: the fastest, cheapest graph DB

The deepfake detector that goes where you browse

The podcast player for sharing.

A free reader that answers your questions and reads aloud

Build agents that heal themselves.

The Best AI Browser Build both for You and your Agents.

Control Every AI Request

OpenAI's most capable model for end-to-end work

Learn AI skills for your specific role and team

Your replies ready with your work's full context

Your Slack org chart, built by everyone in it.

Live commute times in your menu bar

Your honest journal

Lets your AI agents debug production without redeploying

Review code with your coding agents before committing

On-policy distillation (OPD) offers dense token-level supervision as an alternative to the sparse outcome-level advantages of reinforcement learning with verifiable rewards (RLVR). However, the teacher scores student-generated trajectories that are inherently off-policy for it, so the reliability of its supervision, and hence the source of the student's improvement, remains unclear. We quantitatively analyze teacher supervision during OPD training and find substantial noise whose prevalence increases with teacher scale. Surprisingly, the student policy is insensitive to such noise, converging to comparable performance regardless of whether noisy supervision is retained or removed. Does OPD distill at all? By analyzing what drives its gains, we find that learning concentrates on low log-probability tokens, and using a single fixed negative advantage matches the performance of teacher-provided ones. This suggests that OPD works largely by suppressing low log-probability tokens, which requires no teacher. These findings motivate On-Policy Self-Adaptation (OPSA), a supervision-free method using entropy-adaptive negative advantages. It assigns stronger learning signals to high-entropy positions, suppressing tail tokens, and evenly redistributing probability mass among head tokens. Compared with the base Qwen3-1.7B, OPSA improves Avg@32 by 35.41 points on AIME24, corresponding to a 263\% relative gain, and more than doubles Pass@32 across all three benchmarks. It also outperforms OPD by 16.77 points in Avg@32 on AIME24. Extensive experiments and analyses across model families and tasks further demonstrate its effectiveness and generalizability.

Recent video generators often omit audio or synthesize it in a separate stage, limiting reciprocal modeling of visual dynamics and acoustic events. We present DreamX-Creator 1.0, a compact native joint audio-video generation system centered on a 7B generator. Conditioned on a first frame and a text prompt, the generator jointly denoises modality-specialized audio and video streams. The streams are processed independently in the first half of the network and coupled in the latter half through Gated Cross-Modal Attention, whose token- and head-wise output gates modulate each active cross-modal attention-head output. A unified Audio-Video Data System constructs and filters temporally coherent clips, produces structured multimodal annotations, and organizes clips into capability-oriented data pools. Progressive Joint Training comprises two audio-video pre-training stages followed by High-Quality Finetuning. Audio-Video Reinforcement Learning further post-trains the generator with Modality-Aware Multimodal Feedback that routes video-, audio-, and cross-modal feedback to the corresponding streams. For high-resolution output, our Autoregressive 1-Step 2K Refinement pipeline adapts a bidirectional multi-step teacher into an autoregressive multi-step refiner and distills it into a student requiring one denoising evaluation per temporal chunk. Overall, DreamX-Creator 1.0 achieves native, synchronized audio-video generation with performance competitive with state-of-the-art open-source systems. By releasing our compact 7B generator and 2K Refiner, we seek to democratize native audio-video generation and provide an accessible foundation for future research in unified audio-video generative modeling.

Composable scene modeling aims to recover a real indoor scene as complete, editable object assets arranged as observed, giving robot simulation and embodied AI a simulation-ready replica of the real environment whose objects can be manipulated individually. Existing pipelines decompose the task into three steps---parse the observations into instances, generate an asset for each, and place each asset back---but every step presumes an input that a cluttered capture rarely provides: accurate instance geometry, unoccluded views, and assets that accurately match the observations. We propose Lucida, which keeps this order but redistributes the requirements, so each step consumes only what a real capture reliably provides and precision is reached at the end of the pipeline rather than demanded at its start. Lucida parses the video into a scene graph whose nodes carry per-instance multi-view evidence, generates a complete asset for each instance from its evidence, and places assets with GizmoAct, a VLM policy that casts placement as multi-turn GUI interaction, manipulating the object's gizmo in a closed loop and deciding itself when alignment is reached. Across scene-level 3D object detection, object pose estimation, and scene reconstruction, Lucida improves mAP over Boxer by 69% on R2S-Scene, raises [email protected] from 57.8% to 83.4% on CA-1M, and increases scene F-Score from 0.794 for SAM3D to 0.924.

Latent generative models typically follow a two-stage pipeline, training a variational autoencoder for reconstruction and then a generative model on the frozen latent space. Since reconstruction-optimized latents are not necessarily generation-friendly, jointly training both models is an appealing alternative. However, direct end-to-end training remains challenging, as it is prone to latent collapse and faces a generation-reconstruction conflict. We revisit this problem by analyzing how different objectives shape the latent space and identify two key insights. First, the entropy term in the Kullback-Leibler divergence objective is essential for preventing collapse: reconstruction and prior fitting tend to shrink the posterior, while entropy preserves non-degenerate latent uncertainty. Second, reconstruction and generation exhibit asymmetric learning dynamics: reconstruction is fast and strongly supervised, whereas generation is slower and harder to optimize. Based on these insights, we achieve the first direct end-to-end training without latent collapse and propose GenFirst, a simple generation-before-reconstruction strategy. The generative objective first shapes the latent space under weak reconstruction pressure, after which reconstruction is progressively strengthened to recover visual details. We validate GenFirst with continuous autoregressive priors with exact likelihoods and SiT priors with implicit likelihoods. With our end-to-end objective and GenFirst, SiT achieves a gFID of 0.97 with CFG and 1.45 without CFG on ImageNet-256, while MMDiT reaches a GenEval score of 0.90 on text-to-image generation. Beyond image generation, we extend the framework to shared visual latents for generation and representation learning, and to continuous unified text-image generation. These results demonstrate the generality of stable end-to-end latent learning across generative priors and modalities.

AI tutors are most useful when they adapt to each student's strengths, weaknesses, and preferred guidance, but evidence about which guidance works for which student is sparse, slow, and costly to collect from real learners. Student simulators can provide this signal as a proxy, yet existing approaches are limited: state-tracking models fit student behavior but struggle to process explanations or corrections, while LLM role-play follows guidance fluently but does not reliably match the competence of the student being imitated. We present StudentSim, a training framework that turns sparse per-student data into individualized simulators through pooled training followed by per-student specialization. The resulting simulators both mirror a student's own responses and update them under tutor guidance. We also introduce StudentSimEval, a standardized protocol covering 60 students across chess, second-language English writing, and mathematics, using public learner datasets with de-identified records shared for research. StudentSimEval measures behavioral fidelity (F), or how well a simulator matches a student's responses, and guidance responsiveness (R), or how readily it updates under tutor guidance, with all methods fit and evaluated on the same records. Across all three domains, StudentSim outperforms GPT-5.4 on both metrics. In chess, StudentSim reaches F=0.51 and R=0.91, compared with 0.23 and 0.72 for GPT-5.4 and 0.45 and 0.27 for Maia2. As a proof of concept, using StudentSim as a reward model for tutor reinforcement learning produces a chess tutor that expert humans rate as more accurate, better-guided, and more personalized than a no-RL baseline and a tutor trained against a GPT-5.4 simulator reward. Code is available at https://github.com/microsoft/StudentSim.

We present Qwen-Drive-1.0, an initial step towards a vision-language foundation model for autonomous driving. Qwen-Drive-1.0 retains the architecture of the pretrained vision-language model (VLM) and integrates 3D perception, visual question answering, and motion planning within a unified framework. An external bird's-eye-view (BEV) perception head jointly performs 3D object detection, semantic occupancy prediction, and BEV map segmentation. It serves as a probe of the 3D information accessible from the shared representations and provides an explicit, inspectable interface to 3D scene structure. A Planning Expert conditions on shared VLM representations to generate future ego trajectories. A staged training recipe combines driving supervision with general-purpose vision-language data to acquire driving-specific competence while helping preserve broad visual understanding and instruction-following capabilities. Experiments demonstrate strong 3D perception and driving scene understanding while largely preserving general vision-language capability. Comprehensive evaluations across open-loop, pseudo-closed-loop, and closed-loop settings further show highly competitive motion-planning performance.

Looped Transformers increase effective depth by iterating a shared block of layers, but most evaluations compare at fixed model size, conflating architectural advantage with extra FLOPs. We study looping on Mixture-of-Experts Transformers while closely matching per-token FLOPs, total non-embedding parameters, and KV cache. Through a series of ablations, we arrive at a recipe we call SMELT (Sparse MoE Transformer, middle layers Loop Twice), which loops the middle half of layers twice while matching the unlooped Baseline on all three budgets. We scale SMELT across four sizes up to 54B non-embedding parameters and fit a separate Chinchilla-style scaling law for each architecture. SMELT's loss drops faster with compute, saving 6.8--18.0\% of training FLOPs on the compute-optimal frontier. The advantage transfers to downstream benchmarks beyond what validation loss predicts, is largest on Code, and grows with sample length and the number of in-context examples. Mechanistic analysis shows that the second visit reduces the attention sink and redirects mass toward content-relevant tokens, an inductive bias that may underlie the observed performance gains. These results show that looping can improve Transformers even under budget matching, offering a practical recipe that turns depth reuse into measurable gains.

Multimodal GUI agents have emerged as a promising paradigm for digital task automation, yet transitioning from benchmark-oriented models to dependable real-world applications remains challenging due to limited environment coverage, brittle task construction, and unreliable reward verification. In this work, we present UI-Venus-2, a general-purpose foundation GUI agent designed to operate across mobile, web, and desktop environments through a unified closed-loop reasoning-action framework. To bridge the gap toward practical deployment, we jointly scale three critical dimensions: (1) Environments, expanding coverage to more than 170 multilingual mobile apps and native desktop operating systems; (2) Tasks, employing a deep-research pipeline for function-grounded instruction generation; and (3) Verification, adopting trace-level and sample-level evaluators with visual keypoints and multi-model voting to ensure reliable RL signals for training. Furthermore, we integrate safety-aware mechanisms to ensure controlled execution of consequential actions. By offering a capable, efficient, and open-source foundation, UI-Venus-2 advances the field toward more generalizable, verifiable, and self-reflective agents for real-world applications.

Autonomous agents are beginning to carry out machine-learning (ML) research end to end. These agents combine a model backbone with a harness for planning, execution, memory, and verification, but this architecture still leaves domain-specific know-how outside the agent. We call this missing layer operational knowledge, the know-how that separates knowing a method from making it work. That knowledge is not absent from the field. It appears in repositories and papers, but in forms written for human readers and too large to load during a task. Once distilled into compact, verified skills, this knowledge can be reused across tasks rather than rediscovered during each run. We present DisCo, a skill-powered research agent that creates skills and uses them during research. Its distillation runs in two complementary forms: task-agnostic, condensing the field's widely used repositories into reusable skills, and task-oriented, producing the skills a concrete task calls for. The former, applied across the open ecosystem, yields the AREX-Skill Library, with 5,000+ verified skills distilled from 1,000 widely used ML repositories and organized into 20 areas and 178 capability families. With the GPT-5.5 backbone, research harness, and downstream execution budget held fixed, the skill-equipped research agent scores 134.3% higher on MLE-bench, 34.4% higher on PaperBench, 9.2% higher on FrontierCS, and 14.0% higher on PassNet than the same agent without skills. These gains come from adding distilled operating context under that fixed setup.

As agents move from research prototypes to deployed tools, their capability increasingly depends on model-external execution infrastructure, commonly termed the agent harness. Changing this harness while holding model weights fixed can substantially alter task performance. Current agent evaluations typically report downstream performance under a chosen harness, leaving a model's ability to develop the harness itself comparatively underexplored. We introduce HarnessDev, a benchmark that shifts the unit of evaluation from task outputs to runnable infrastructure. HarnessDev covers two stages. In Creation, the agent starts from a minimal seed and a small number of cases, then builds a complete execution system. In Evolution, it starts from its own created harness and iteratively revises it using downstream execution feedback, with the goal of improving benchmark performance. We then evaluate each constructed harness on capability (task success on held-out benchmarks) and efficiency (execution-token cost). The reported Creation results cover six creator LLMs, four domains, and five downstream benchmarks totaling 2,207 unique downstream instances, with hidden evaluation tasks withheld from development. We find that generated harnesses remain substantially behind mature human-engineered references on code and on search and research, while matching or exceeding the selected references on writing and machine-learning experimentation, with large variation in execution cost. Evolution produces some performance gains, but they are unstable and transfer only partially to held-out tasks. Experiments with a fixed runtime model further show that the gains depend strongly on the model executing the harness, indicating limited transfer across models.

Many important forms of human learning begin with a vague goal, such as "become a better physicist" or "improve at research." Learners must interpret the goal, identify capability gaps, decide how to learn, and determine whether they have actually improved. In contrast, existing work on LLM self-evolution typically begins with tasks and evaluation metrics specified by humans, reducing self-evolution to optimizing an explicit objective rather than deciding what and how to learn. We introduce ASPIRE, a benchmark for vague-goal-driven self-evolution. ASPIRE provides only a natural-language capability goal while downstream evaluation tasks remain hidden. The agent must operationalize the goal by choosing data and update methods, constructing training and validation signals, and deciding when to evaluate. ASPIRE supports both model-weight and agent-harness evolution in a unified interactive environment and evaluates the resulting systems on a hidden, expert-authored set of 520 items spanning six goals. Our experiments show that vague goals redirect search effort toward goal interpretation. Current agents routinely complete training and harness-editing loops, but weight-level gains remain sparse and unstable, and the strongest evolved harness remains below the engineered Qwen-Agent reference. Agents often train on mismatched data and trust narrow self-evaluations, so local gains fail to transfer to hidden evaluation and continued search and training can erase earlier improvements.

We introduce SolarWM, a fully open foundation for building interactive video world models from data preparation through long-horizon inference. Training across heterogeneous data sources and video backbones is challenging: datasets differ in temporal scale, camera geometry, visual quality, motion, and captioning styles, while video generators use distinct representations and architectures. Naive data mixing and model-specific implementations therefore produce inconsistent supervision and make results difficult to reproduce and compare. SolarWM addresses this coupling with a reconfigurable multi-source data engine and a backbone-native adaptation framework. The engine converts 1.43 million canonical clips from 10 datasets into a unified, frame-aligned contract covering visual observations, metric camera geometry, captions, quality metadata, selection decisions, and provenance, while decoupling source processing from mixture construction. Under shared camera-conditioning, training, and inference interfaces, we instantiate four 5B--33B models based on Wan2.2, LTX-2.5, and MiniMax-H3 while preserving their native representations and objectives. A unified three-stage recipe combines bidirectional adaptation, teacher-forced autoregressive initialization, and distribution matching distillation. The resulting causal models enable real-time interaction over rollouts ranging from minutes to hours after being trained on only 5s sequences. By releasing the resulting data, pipeline, recipes, weights, and framework, SolarWM provides a reproducible and extensible foundation for interactive world-model research.

Many recurring text functions are easy to describe but difficult to implement with rules, while calling a large remote model for every input introduces repeated cost, latency, and dependency on a provider. We present compile by training, which turns a natural-language specification into a reusable neural function. At compile time, teacher models generate task-specific examples that are used to train a small adapter for a compact interpreter. The resulting function runs without the teachers and can be stored, versioned, and composed like ordinary software. On FuzzyBench-Hard, a subset on which the Program-as-Weights fast compiler produced no exact matches, compile by training reaches 83.6% semantic accuracy. This higher accuracy comes with a higher compile-time cost: roughly a minute rather than seconds for the fast compiler. We deploy the compiler in a public interactive service and demonstrate compiled functions in a multi-site website helper, a language-controlled 3D avatar, and a bidirectional English-Claudish translator.

As terminal-based code agents become prevalent, agent trajectories have accumulated at scale, while realistic, executable environments remain scarce. However, environments are what agent post-training actually requires: each can be re-queried into many verifiable tasks and provides execution feedback, whereas a trajectory is a single frozen demonstration. Rather than generating environments from scratch, we observe that the tool-execution history in existing trajectories exposes the structure and contents of the environments in which they ran, making it possible to reconstruct those environments from the trajectories themselves. Thus, we introduce Terminal-Universe, a framework which turns each trajectory into a reusable environment and explores it for synthesizing new tasks and continued interactions. Specifically, Terminal-Universe replays the file operations recorded in a trajectory to restore each file before the agent modified it, yielding a partial workspace; a completion agent then supplies the missing files and dependencies. On this recovered workspace, we both reconstruct the original intent task and synthesize entirely new ones. Besides, we also scale the tasks along two complementary axes: breadth and depth. For breadth, we mine directional dependency relations between related environments and synthesize cross-workspace queries spanning multiple codebases, as developers routinely do in real-world development. For depth, we extend the initial single-turn query into a multi-round session that captures iterative user feedback and requirement refinement via a user agent. Applied to public terminal agent trajectories, Terminal-Universe produces 37.3k task-sufficient environments. Supervised fine-tuning of Qwen3.5-27B on this corpus improves single-round performance on Terminal-Bench 2.1 by 11.9 points and multi-round performance on EvoCode-Bench v2 MT@4 by 13.8 points.

We introduce LLaDA-Image, a unified framework that pairs a 6B Diffusion Transformer (DiT) trained from scratch with a frozen vision-language understanding module built on the LLaDA2.0-Mini diffusion language model backbone. Instead of relying heavily on paired image-text data from the beginning, we first build a strong visual generative prior through image-only pre-training and mid-training. The generation pipeline comprises 220M samples, 98 of which are real images. For efficient and scalable optimization, we use parameter-free RMSNorm throughout the DiT together with the Muon optimizer. The resulting unified model produces highly photorealistic images while accurately following fine-grained editing instructions. We further distill LLaDA-Image into LLaDA-Image-Turbo, enabling fast inference in 2-4 sampling steps. On Qwen-Image-Bench, LLaDA-Image achieves overall scores of 53.53 and 53.38 on the English and Chinese tracks, respectively, setting a new state-of-the-art among open-source models on both tracks. To support further research on capable and efficient generative models, we release our model weights, training code, and detailed recipes.

Large language models offer broad capabilities, but adapting them to evolving domains, tools, and requirements often entails repeated post-training. Autonomous systems automate parts of this process by proposing updates, training candidates, and using evaluation feedback to select subsequent proposals. As evidence accumulates, a central problem emerges: which past update evidence remains actionable after subsequent training has changed the parent model? An update's effect depends on its parent, data, and training stage. Treating past success as context-free permission can waste compute. If the resulting child is promoted, it can also degrade the subsequent training trajectory. We formulate this problem as conditional experience transfer and introduce Boundary-Calibrated Intervention Transfer (BCIT), a method that authorizes experience reuse before weight-changing training. BCIT binds an observed effect to its source context, checks applicability conditions, vetoes candidates with named hard conflicts, and obtains current-state evidence through a bounded training trial when needed. Fully trained candidates still face a shared adoption rule, and only observed events extend memory. On one 4B model adapted across finance reasoning, text-to-SQL, and function calling, candidate updates exhibit heterogeneous target and retention effects across the evaluated contexts. Under matched candidates, evidence, and compute, BCIT authorizes fewer harmful updates and attains higher equal-budget final-model quality than the evaluated alternatives. These results support treating experience authorization as a distinct problem in autonomous post-training.

Large language models achieve superior performance on tasks that require extended reasoning, but long chains of thought make the KV cache a severe memory bottleneck. Existing KV cache compression methods share one paradigm: score each cached token by some estimate of how much it will matter later, and keep the top-scoring ones. We show that the selection signal contributes almost nothing. Random Attention keeps the prompt and evicts uniformly at random within each attention head, computing no score at all; across four models and six reasoning tasks it matches the strongest prior evictor while serving 32-43% higher throughput than it in vLLM deployment. Controlled experiments explain this by showing that 1) the prompt is the fragile part of the cache, and most of the gap between selectors is just whether their selection signal happened to keep it; 2) the reasoning trace protects itself against eviction with redundancy at two levels, in the text (the model restates what it still needs as it works) and across attention heads (each keeps its own copy of the trace), so once the prompt is safe, a random draw retains enough copies of what the model still needs, and no score is required to pick them. Our code is publicly available at https://github.com/SalesforceAIResearch/Random-Attention.

Multi-agent LLM systems commonly use an orchestrator to decompose a task for a team of workers and then improve through textual reflection. Despite strong empirical results, these systems lack a unified account of coordination, memory improvement, and the role of external verification. We model orchestrator-worker interaction as a bilevel coordination game: under bounded coupling, the workers' local-update game is an approximate potential game whose equilibrium slack is controlled by decomposition quality. We then analyse reflection as stochastic movement over semantic memory states. For free-form reflection, we derive a finite-time upper bound, prove worst-case tightness, and give a positive lower bound under a falsifiable persistent-harm condition. We further prove an information-theoretic impossibility result: no gate that observes only the generated transcript can improve uniformly over text-indistinguishable environments, whereas an environment-grounded gate can. Motivated by this separation, we introduce Stochastic Reflective Memory Ascent (SRMA), which accepts a candidate memory only after a grounded evaluation risk strictly decreases. Under calibration and non-degenerate corrective mass, SRMA converges exactly, geometrically or polynomially; matching constructions show that both rate regimes are order-tight. We also provide confidence gating for stochastic evaluation and re-anchoring guarantees for piecewise-stationary environments. Experiments instantiate these objects with environment-grounded metrics and test the predicted coordination and drift laws. On 500 SWE-bench instances, the complete Kimi-based system resolves 72.2% versus a 70.8% public mini-SWE-agent reference. Code: https://github.com/YihangChen9/Bilevel-Coordinated-Reflection

We present Iris-mini and Iris-pro, two search agents trained at the 35B-A3B and 397B-A17B scales, together with the data pipeline and training recipe behind them. Tasks are reverse-constructed from the hyperlink structure of a web corpus: we author multi-hop chains over an entity graph distilled from a seed page and its out-links, rewrite every non-answer entity into a descriptive reference so that no clue can be resolved by string matching, and admit only questions that a reference model fails closed-book yet solves once the supporting evidence is supplied. These questions are then turned into trajectories, which are filtered at both the trajectory and the turn level before SFT. The policy is then optimized by RL against live search, with the reward judge and the observation summarizer served inside the training cluster, and with over-long rollouts interrupted at the request level and resumed from their committed prefix at the next step. We alternate the two stages in a procedure we call SFT-RL climbing, returning the hardest solved and most efficient rollouts of each RL round to the next supervised pass. Because inference-time context management is worth more on these benchmarks than most reported differences between systems, we evaluate every benchmark both with and without it, holding the tool set, the context limit, and the judge fixed. All results come from a single ReAct agent, with no sub-agents and no test-time verification. With management enabled, on BrowseComp, BrowseComp-ZH, DeepSearchQA, and HLE the two models reach 82.2/84.8/86.9/52.3 and 88.6/85.1/92.9/56.4, the strongest overall results among open-source search agents in their respective parameter ranges. We plan to release the model weights together with the complete recipe for data construction, training, and evaluation.

An avatar that holds a conversation should decide what to say and to move while saying it, yet these abilities live in separate model families: spoken dialogue models produce speech without motion, and co-speech motion models produce motion only from audio handed to them. The standard remedy is a cascade that first generates the spoken response and then runs a motion model over the finished audio, which requires a second full inference pass and precludes any joint optimisation between the two. We present Motion-Omni, an end-to-end framework in which a spoken dialogue model natively outputs explicit facial expression together with hand, upper-body and lower-body motion, generated directly from the hidden states that produce the speech. Joint training is not optional here: with the speech pathway frozen, motion remains misaligned with the audio, and co-adapting the LLM, Speech Generator and Motion Generator under both objectives is what recovers alignment while retaining spoken-dialogue ability. Supervision comes from a scalable, model-agnostic pipeline that pseudo-labels consistent-voice speech responses with a replaceable motion teacher, yielding 422,856 quality-ranked pairs (1,402 hours). We further release SwDA-500 and, to our knowledge, the first public evaluation protocol for stochastic open-ended full-body spoken dialogue, matching audio across motion systems while unifying rendering, automatic metrics, human evaluation, and latency measurement. Instantiated with a Qwen2.5-7B-Instruct backbone, Motion-Omni-Q7 matches the same-audio teacher cascade to within 2% on reference-free motion metrics while responding 5.4 x faster (RTF=0.78, faster than real time), surpasses all non-teacher cascades on beat correlation and diversity, and reaches a 2.62% word error rate, the lowest among the omni-modal systems compared.

Ryan Whitwam / Ars Technica : Google rolls out its September Android Drop, with remembered items in Find Hub, Guided vision in Gemini Live, Motion Assist to reduce motion sickness, and more —  Let's be frank: Some of Google's recent Android feature Drops have been duds, offering little more than an expansion of Gemini summaries and chat functions.

Mark Gurman / Bloomberg : Filing: Apple says John Ternus will get a compensation package worth ~$58M in FY 2027, while Tim Cook's role as executive chairman will pay ~$47M —  Apple Inc. said that new Chief Executive Officer John Ternus will get a compensation package worth about $58 million in fiscal 2027 …

Erin Woo / Wall Street Journal : Sources: Google plans to release Gemini 3.8 Flash as soon as Wednesday; Gemini 4 has done well on pre-training evals but still needs to complete post-training —  Internal tests of Gemini 3.8 Flash show progress in an area where the company has lagged behind Anthropic and OpenAI.

Reuters : The US urged G20 members to avoid writing entirely new AI regulations, and instead focus on writing rules for “novel” situations that involve the tech —  The U.S. pressed G20 members on Tuesday to take a hands-off approach to AI regulation and avoid creating new rules …

Ryan Gould / Bloomberg : Sources: Thoma Bravo-backed Proofpoint is in talks to acquire cybersecurity company Varonis; Varonis rose 10.4% on Wednesday, giving it a market value of ~$5.4B —  Thoma Bravo-backed Proofpoint Inc. is in talks to acquire cybersecurity company Varonis Systems Inc., according to people familiar with the matter.

@artificialanlys : Muse Spark 1.3 with max reasoning, in limited preview for partners, scores 62 on the Artificial Analysis Intelligence Index, behind only Fable 5.1 and Opus 5 —  Meta has released Muse Spark 1.3, their fourth Muse Spark model release in five months. Muse Spark 1.3 (max), which is in limited preview for Meta's partners, scores 62 on the Artificial Analysis Intelligence Index, behind only Claude Fable 5.1 and Claude Opus 5. The variant available now, Muse Spa...

Dina Bass / Bloomberg : HPE reports Q3 revenue up 34% YoY to $12.2B, vs. $11.9B est., forecasts FY 2026 and FY 2027 sales growth above est., signs a $3.5B server deal with a cloud firm —  Hewlett Packard Enterprise Co. boosted its sales outlook for this fiscal year and next, saying artificial intelligence workloads …

Jordan Novet / CNBC : Broadcom reports Q3 revenue up 86% YoY to $29.59B, vs. $29.36B est., AI semiconductor revenue up 221% to $16.7B, and forecasts Q4 revenue below estimates —  Broadcom shares dropped 5% in extended trading on Wednesday after the chipmaker issued disappointing guidance for the current quarter.

Axios : Sources: PayPal has paused plans to sell its VC portfolio for $900M-$1B after bids came in lower than hoped, months after deciding to wind down PayPal Ventures —  PayPal has paused plans to sell its venture capital portfolio for around $900 million to $1 billion, Axios Pro has learned from multiple sources.

Subrat Patnaik / Bloomberg : Oura files for a US IPO, reporting a $924.3M net loss on $1.21B in revenue for the nine months ended June 30, vs. a $182.8M net loss on $697.6M a year earlier —  Health and fitness ring-maker Oura Inc. filed for an initial public offering, showing soaring revenue as well as increasing losses.

Annie Palmer / CNBC : Adobe names Customer Experience Orchestration president Anil Chakravarthy as CEO, effective December 1, replacing Shantanu Narayen, who becomes executive chair —  Adobe on Thursday named Anil Chakravarthy as its next CEO, succeeding Shantanu Narayen, who announced he would step down earlier this year.

Sam Sabin / Axios : OpenAI commits $1B in subsidized model access, training, support, and partnerships to a new initiative aimed at protecting essential services around the world —  - The $1 billion will come in the form of expanded subsidized access to models, training, technical support and partnerships for participating organizations.

Chase DiFeliciantonio / Politico : California AG Rob Bonta is investigating OpenAI over the Hugging Face hack in July, after more than a dozen states joined Alabama in its investigation —  SAN FRANCISCO — California Attorney General Rob Bonta is investigating OpenAI over the recent hack its programs carried out on their own …

Jonathan Randles / Bloomberg : Filing: AI training data startup Micro1 offers to pay $12.5M for Spirit Airlines' data; the offer faces hurdles as Spirit already has a $10M deal with Google —  An artificial intelligence startup is attempting to wrest away a vast trove of Spirit Aviation Holdings Inc.'s business records from Google LLC.

Anthropic : Anthropic says Claude worked “largely autonomously” over 11 days to formalize the proof of Fermat's Last Theorem in the Lean programming language —  We are sharing the first complete computer-checked proof of Fermat's Last Theorem.  Claude worked largely autonomously over 11 days …

Bloomberg : Sources: London-based AI infrastructure startup Nscale is in talks to raise as much as $3.5B in financing, including $2B from Nvidia, ahead of a planned IPO —  Nscale, a cloud computing firm focused on artificial intelligence, is in talks with potential investors to raise as much as $3.5 billion …

Todd Bishop / GeekWire : The Seattle Times and Newsday sue OpenAI and Microsoft, alleging the companies trained AI on their journalism; Microsoft and OpenAI are funders of Seattle Times —  Microsoft was sued Friday by the parent company of its hometown daily newspaper, The Seattle Times Co., which joined with Newsday …

ABC News : Trump admin debuts arcade.gov with a deportation-themed Snake clone and a Tetris-like wall-building clone; Tetris says it takes IP infringement “very seriously” —  In “Build the Wall,” players aim to “protect the border” from zombies.  —  Isabella Murray, Nicholas Kerr, and Michelle Stoddart

Jean Eaglesham / Wall Street Journal : Swiss Re: global premiums for insuring data centers will likely reach $20B-$30B/year by 2030; ~40% of US data-center capacity is located in tornado-prone areas —  AI hyperscale data centers are exposed to a dizzying array of potential threats during construction and operation

Miranda Murray / Reuters : Berlin is reviewing Rhysida's 5.79TB release of state data after refusing to pay a ransom; files reportedly include national defense and threat response plans —  Berlin's state government said on Saturday it was reviewing with the highest intensity a trove of stolen data published by a ransomware group …

周二公布的一项民调显示，七成韩国民众表示支持出台限制青少年使用社媒的政策。这项民意调查访问了 1000 名年龄在 14-58 岁之间的受访者。调查结果显示，70.7% 的受访者支持，29.3% 的受访者反对。占总调查人数五分之一的青少年受访者中，59% 反对，41% 支持。当被问及实施此类限制的合适年龄时，16.8% 选择了 15 岁，14.9% 选择了 18 岁，13.1% 选择了 11 岁及以下。大多数受访者表示，即使出台此类政策，在限制青少年使用社媒方面仍然存在局限性，青少年用户可能会盗用他人账号或转向其它不受限制的平台，因此 59.8% 的受访者认为，平台应采取更多措施营造安全的社媒使用环境。

8 月 28 日 20:57 UTC 左右，一个不相关网络 BGP 路由通告了 Softaculous 使用的 Hetzner IP 段，将部分原本发送到 Softaculous 系统的流量劫持到攻击者控制的服务器。Hetzner 是 Softaculous 的上游基础设施供应商，而 Softaculous 则是一家为 Web 托管服务商提供软件的公司，它的 Virtualizor 控制面板被管理员用于部署和管理 VPS。这次 BGP 路由劫持影响了 Virtualizo 更新服务器以及客户和计费网站。攻击者还从 Let's Encrypt CA 获取了有效的 TLS 证书，Let's Encrypt 的自动域名所有权验证也被劫持到了攻击者控制的 IP。Softaculous 于 8 月 29 日 08:50 UTC 向 Hetzner 报告了事件，Hetzner 随后通过发布相同的路由通告遏制了问题。但攻击者于 20:00 UTC 再次了长达 10 小时的路由劫持。8 月 30 日 05:50-06:10 UTC 路由通告被撤回，劫持停止。Softaculous 建议在攻击期间登陆过的用户立即重置密码，以及重置所有重用该密码的账户。同一时间段内输入过银行卡信息的客户也应检查其账单。攻击者在此期间推送了一个恶意的 Virtualizor 更新包，它建议所有 Virtualizor 用户检查其服务器并轮换凭证。

为弄清动物进入冬眠时大脑的变化，研究人员首先在实验室中诱导叙利亚仓鼠冬眠。两个月里，他们把动物笼舍中开灯的时间缩短，以模拟秋季。然后在接下来的两个月中，将温度降至约4摄氏度，以模拟冬季。 在人造冬季中，动物开始冬眠——体温下降，“在窝里缩成一团”。冬眠持续2到8周，其间仓鼠睡眠状态在微觉醒和深度蛰伏之间循环。团队收集了刚进入深度蛰伏的仓鼠的大脑，并将其与刚从蛰伏中短暂醒来或完全未冬眠的仓鼠的大脑进行比较。结果发现，一种名为Fos的蛋白质高水平表达，表明下丘脑视前区存在活动。该区域参与调节体温、睡眠和其他重要功能。研究人员发现，仓鼠冬眠中活跃的POA神经元特定亚群，似乎与小鼠蛰伏状态中鉴定出的神经元相同。抑制这些神经元会延迟仓鼠重新进入蛰伏，而激活这些神经元则引发仓鼠筑巢行为并使其体温下降。这是体温虽不如自然冬眠时那么低，但也远低于平时，仅13摄氏度。在小鼠中激活这些细胞也能使体温降低，但幅度较小。这组POA细胞可能是进化遗留下来的“关闭键”，使早期温血哺乳动物能够降低维持体温的高能量成本。

欧盟委员会周一表示，OpenAI 的 ChatGPT 将需要遵守更严格的欧盟法规，否则将面临罚款。聊天机器人 ChatGPT、社媒论坛 Reddit 和游戏平台 Roblox 被欧盟网络安全法规《Digital Services Act（DSA）》归类为“超大型在线平台”。该认定意味着这些服务面临额外的义务，如删除非法内容、保护未成年人的隐私和安全，如果未能遵守规定，将面临最高全球收入 6% 的罚款。欧盟的决定标志着 DSA 的适用范围进一步扩大到生成式 AI 领域。此前 X 的 AI 聊天机器人 Grok 已因违反 DSA 而受到调查。这三大服务在欧盟的月活用户数都已超过 4500 万，达到了 DSA 规定的加强审查门槛。它们需要在 12 月底前履行额外义务。

全世界正在快速发展太阳能这一廉价电力来源，但各地发展并不均衡。以欧洲为例，除了罗马尼亚，其它欧洲国家几乎没有增长，原因是欧洲大陆电网的局限性。发电高峰期间电网拥堵，储能不足，导致能源浪费、负电价，甚至出现电网故障和灾难性停电风险，欧洲各国正竞相增加电网储能容量以解决这些问题。罗马尼亚太阳能装机量在 2025 年增长了 45%，该国在太阳能建设早期就对电池储能进行了投资。除此之外，巴基斯坦的屋顶太阳能和电池安装速度超过了其他任何市场，巴西、智利、萨尔瓦多、摩洛哥、肯尼亚和纳米比亚在清洁能源转型方面都已超越了美国。

根据发表在《Frontiers in Public Health》期刊上的一项研究，全世界每年有近五成农民因农药中毒。研究发现每年约有 4.02-4.33 亿农民和农场工人因意外急性农药中毒。全球农业人口约为 9.34 亿，这意味着每年约有 46% 的农民农药中毒。中毒人数最多的地区是南亚，其次是东南亚和东非。中毒率最高的国家是西非布基纳法索，该国近 84 %的农民和农场工人曾遭受农药中毒。研究还发现，每年约有 11,000 人死于农药中毒，其中近 60% 发生在印度。研究指出，农药的使用量仍在持续增长，2023 年的使用量达到 380 万吨，比 1990 年翻了一番。

Steam 平台一周内上架的游戏数量首次突破 700 款，但大部分游戏无人问津。根据 ICO Partners CEO Thomas Bidaux 提供的数据，上周共有 720 款游戏在 Steam 平台上架。其中 520 款游戏的评论数在 0-9 条之间，190 款游戏的评论数超过 10 条，95 款游戏的评论数超过 50 条，62 款游戏超过 100 条，10 款游戏的评论数超过 1000 条。数据存在重叠，评论数超过 1000 条的游戏会被同时计入 50+ 和 100+ 类别中。Steam 平台在 2026 年上半年上架了约 12000 款游戏，同比增长 19%，游戏销售收入达到了 111 亿美元。

根据稳定版内核维护者 Greg Kroah-Hartman 制作的幻灯片，Linux 7.x 系列的 CVE 数量已从 7.1 的逾千增加到 7.2 的逾 1500，按照这一趋势，下个版本 7.3 将会超过 2000。这一切并非是因为 Linux 内核安全性变差，而是因为 AI 辅助安全检测工具对内核源代码的自动扫描发现了大量 bug，大部分 bug 并不严重属于低危级别。AI 工具产生了大量报告，而要从这些报告中发现有用的信息需要维护者耗费大量精力和时间。内核网络系统的维护者 Jakub Kicinski 表示他们有点不堪重负了。为了减少旧代码的 bug 报告，内核维护者们开始移除大量基本上无人使用的驱动代码。Linux 7.3 移除了旧的 SGI 和 IBM 驱动代码，此举旨在减少维护成本，因为这些历史悠久的代码被 AI 工具发现了大量 bug，而维护者有义务调查和修复这些 bug。

根据发表在《Cerebral Cortex》期刊上的一项研究，科学家招募了 18-74 岁的成年人，最后的参与者包括 17 名青年人（18-30 岁）、21 名中年人（50-60 岁）和 23 名老年人（61-74 岁）。每一位参与者都躺在 MRI 扫描仪中完成了一项多步骤的记忆任务。参与者首先接受静息扫描，闭上眼睛保持静止不动；接着在学习阶段观看一张脸与一物体或一场景配对出现，被要求想象此人互动的情景以帮助记住配对信息；最后进行记忆测试。结果显示年龄是记忆表现的重要预测因素。年轻参与者正确识别配对的频率远高于中年或老年参与者，而中年和老年人的表现则相近。老年人表现出更多的混淆错误。老年人的大脑可能并非只是记忆力较弱，他们的大脑会记住过多错误的信息，记忆被扭曲混淆了。

纽约市长 Zohran Mamdani 周三宣布了一项为期一年的新政策，禁止从幼儿园到八年级的公立学校学生使用生成式 AI。纽约市同时还限制低年级学生的屏幕使用时间，为高中生开设 AI 素养课程。禁令将于 2026-2027 学年生效，将影响近 60 万名学生，占全市公立学校学生总数的 2/3。Mamdani 表示，“这项禁令体现了我们对未来负责的承诺，我们将拥抱新技术，但前提是它必须服务于我们的学生。”新政策引入了新的屏幕时间指导方针，包括限制低年级学生一对一使用电子设备，建议 3-5 年级学生每日使用时间上限为 30 分钟，6-8 年级学生每日使用时间上限为 45 分钟。

在任天堂之后，另外两大游戏机制造商微软和索尼都表示他们无法律义务将美国关税退款退给消费者。去年美国在全球大规模征收关税，任天堂等公司据此提高了相关产品的售价。今年 2 月美国最高法院裁决征收关税违法，任天堂等公司随后起诉美国政府要求退还关税。玩家们随后起诉了任天堂、微软和索尼等公司，想要获得部分退还的关税。任天堂表示它无法律义务将美国关税退款退给消费者。索尼和微软都持相同的立场。索尼表示它预计将会获得 5.08 亿美元的关税退款，大部分退款将交给游戏部门。索尼律师本周在一份请求法庭驳回诉讼的动议中表示：“为自愿购买的消费品支付公平的市场价格，不构成法律上可认定的实际损害。”微软律师 8 月 21 日在类似案件中提出了相似的观点。

音频编辑器项目 Audacity 释出了 v4.0。主要变化包括：基于 Qt6 重构了 UI，支持原生高 DPI 渲染；能将 UI 布局保存为 Workspaces；工具栏和面板可移动、停靠、浮动、显示或隐藏；改进了音频片段的处理；新增 .aup4 项目格式；等等。Audacity 4.0 与 Audacity 3.x 不完全兼容，部分 Audacity 3 系列的功能尚未移植到 4.x 版本。

硅谷亿万富翁、Palantir 联合创始人 Peter Thiel 前不久携家人移居了阿根廷，并斥资 1200 万美元购买了一栋豪宅。许多人猜测他可能是为了躲避可能的三次世界大战，阿根廷地处南半球深处，农业资源丰富，是末日避难的理想之地。然而 Peter Thiel 在阿根廷深入参与了当地的政治活动。他与阿根廷总统 Javier Milei 及其亲密顾问进行了多次闭门会谈，之后 Milei 的政府提出了一系列符合 Peter Thiel 利益的提案，其中一项提案要求政府民事部门与国家情报机构共享阿根廷公民个人数据。很多人怀疑 Thiel 的 Palantir 公司可能与阿政府签订了未公开的合同。阿根廷众议院为此邀请 Thiel 和 Milei 前来回答一系列相关问题，但无人出席会议。随后阿根廷人在 Thiel 的豪宅前举行了抗议活动。

巴西科学家分析了 53 条当地捕获的金枪鱼，发现几乎所有鱼都携带了寄生虫。研究团队共统计到 1600 多条寄生虫；肾脏和心脏是唯一没有寄生虫的器官；除此之外寄生虫遍布全身；53 条鱼有 96% 携带了寄生虫；肠道是感染最严重的器官，其次是肌肉组织（即食用部分），然后是胃。巴西研究人员分析的是学名为 Katsuwonus pelamis 的金枪鱼品种鲣鱼，它是巴西最丰富的金枪鱼品种，也是 2022 年全球捕捞量第三大的海洋鱼类——当年全球共捕捞并销售了 310 万吨鲣鱼。

犹他州成为美国第一个要求 VPN 软件验证用户年龄的州。犹他州的年龄验证法律《Online Age Verification Amendments》明确将 VPN 软件纳入监管范围。这项法律规定，任何商业网站，如果逾三分之一内容被认为对未成年人有害，必须实施合理的年龄验证方法才能允许犹他州用户访问。只要用户的实际位置在犹他州内，不管是否使用 VPN 之类的方法隐藏地理位置，都需要验证年龄。违规网站将面临每次违规最高 2,500 美元的罚款。网站也被禁止发布如何使用 VPN 规避年龄验证要求的说明。EFF 认为这项法律可能迫使网站屏蔽所有已知的 VPN 地址流量，还可能违反了美国宪法第一修正案。

今年五月被捕的 Google 安全工程师 Michele Spagnuolo 被控利用内部信息在预测市场 Polymarket 押注歌手 d4vd 成为 2025 年 Google 搜索量最高的人物而获利 120 万美元。他被控犯有商品欺诈、电信欺诈和洗钱罪。 他的律师团队本周递交了一份动议，寻求撤销对他的指控。Spagnuolo 没有直接否认利用内部信息获利，而是宣称他只是在赌博，而赌博行为不是受美国《商品交易法》监管的金融工具，且赌博发生在美国境外，不受美国司法管辖。Spagnuolo 已被 Google 停职。Polymarket 虽然是一家美国公司，但其预测市场在美国被禁，其平台也是由名为 Adventure One QSS 的巴拿马公司运营。Spagnuolo 在 Polymarket 上押注时也不是在美国，而是住在瑞士苏黎世。Spagnuolo 律师团队声称他利用的内部信息对 Google 没有任何价值。

科学家利用欧洲 450,112 名参与者（131,426 名男性 和 318,686 名女性）的数据，评估了红肉、加工肉和白肉与胃腺癌和食管腺癌之间的关联。研究人员对红肉、加工肉和白肉的定义是：红肉包括猪肉、牛肉、小牛肉、羊肉等；加工肉包括火腿、培根、香肠、加工肉块、汉堡包、肉丸、肉酱等；白肉包括鸡肉、鸭肉、鹅肉、兔肉以及其它禽肉。结果显示，每天每摄入 30 克加工肉，胃癌风险增加 9%，食管癌风险增加 13%；每天每摄入 20 克白肉，非贲门胃癌风险增加 12%。

物理学有两大不兼容的理论，其一是爱因斯坦广义相对论，适用于宏观物理领域；其二是量子力学，适用于微观物理领域。爱因斯坦曾尝试将两大理论融合成一个统一的理论，但没有成功。广义相对论的一个核心概念是等效原理，它认为引力和加速度不可区分。等效原理指出，对于自由落体的观察者，引力在局部应该消失。例如在电梯中自由落体的人会体验到失重。根据发表在《Science Advances》上的一项研究，科学家利用冷却到绝对零度附近的铷原子云，将其置于量子叠加态，通过以伽利略名字命名的量子干涉仪，在实验中测试了量子领域的等效原理。结果显示等效原理适用于量子领域，观察结果与引力如何影响量子波的预测相吻合。研究本身并未统一广义相对论和量子力学，但它朝着该目标迈出了一步。

两年前，蒂姆·安德鲁斯疲惫不堪，唯一想做的就是睡觉。糖尿病和高血压毁掉了他的肾脏，而通过过滤血液维持他生命的频繁透析治疗令人精疲力竭。尽管只有 66 岁，他却毫无精力与食欲，虚弱得无法行走，还曾两次心脏病发作。肾移植本可以救他的命，但器官供应短缺。根据美国器官共享联合网络的数据，约有 9 万人在等待肾移植，平均每天有 11 人在等待中去世。而且由于安德鲁斯的血型较为罕见，他获得匹配的人类肾脏的几率尤其渺茫。当他听说麻省总医院布里格姆的医生正在为患者移植转基因猪的肾脏时，他主动联系了他们。“我想，如果我要死了，不如为人类做点事，参与这个实验，”他说。2025 年 1 月 25 日，安德鲁斯接受了转基因猪肾的移植，带着它生活了九个月。今年他成为已知首例在猪肾移植后接受人类供体肾脏的患者，周四发表于《柳叶刀》的一篇论文中描述了这一里程碑式的成就。他的病例为移植医学中一个引人关注的概念提供了初步证据：猪肾虽尚非永久性解决方案，但可以作为通往人类器官移植的“桥梁”。这些猪肾来自经过大量基因编辑的猪，其中一些编辑旨在防止严重的器官排斥反应，还有一些用于灭活猪基因组中的病毒。

一周前，Debian 项目经过投票允许以负责任的方式使用生成式 AI，Debian 项目表示，它既不反对也不支持在软件、包、文档等的开发和维护中使用生成式 AI 工具。但项目也认识到，如果能负责任的使用 AI 工具，将能显著提高贡献者的效率，使他们将有限的时间投入到需要技术专长、判断力、审核和协作的工作中。现在，Android 自由软件应用商店 F-Droid 考虑采用与 Debian 相同的 AI 政策，它的政策提议直接拷贝了 Debian 的政策全文，只是将 Debian 的名字替换为 F-Droid。