AI Daily Briefing

The following article originally appeared on Elevate and is being reposted here with the author’s permission. A software factory is a repeatable loop around software work. If you’re building a software factory, code good enough to ship still needs human taste and ownership. We’ll discuss this including whether you need a factory just yet. If […]

In July, OpenAI revealed that its AI agents had attacked Hugging Face without permission, sparking widespread concerns about AI safety. Since then, a string of similar incidents involving agents from Meta, Anthropic, Google, and other companies has fueled further fears about rogue AI. As disclosures implicating numerous AI models trickled out over the past few months, these seemed like separate incidents. But many share a common source: one specific company tasked with testing the agents. Irregular, an Israeli startup that stress-tests AI models in "high-fidelity research platforms that simulate and monitor real-world AI security scenarios … Read the full story at The Verge.

As AI agents gain more authority to act on their own, recent incidents are exposing gaps in how enterprises monitor, control and intervene when agents cross established boundaries.

Apartment with strange view singled out in government’s fight against deceptive ads – but real estate agent says photo is no fake Get our new political email, free app or daily news podcast The New South Wales government’s crackdown on “AI-doctored” rental listings has hit a wall, after it made an example of a supposedly manipulated ad that turned out to be a genuine photograph of a $5,000 courtyard mural of Bondi beach. In a media release on Friday announcing protections against misleading rental ads, the state government provided a link to a property in Leichhardt, claiming it was an “AI-doctored rental listing in Sydney’s Inner West”. Continue reading...

As the UN warns traditional safeguards are ‘unravelling’, Donald Trump says he will encourage, not restrain, the AI race When the Australian prime minister, Anthony Albanese, sat down for an interview in the heart of Silicon Valley at the weekend he had known for two days that his was the first government known to have been attacked by a rogue AI agent. He didn’t reveal the attack then, but he sounded a warning about the march of AI: “the risk is that AI develops in a way in which humans are no longer in control of what AI is producing.” Continue reading...

Perplexity Research published a new post-training study. It trains a model inside Perplexity Computer on real user sessions, including failed ones. The method pairs rejection sampling fine-tuning with hint-guided self-distillation. In a live A/B test, tool-call failures fell from 2.24% to 1.77% between 2 trained checkpoints. Perplexity team reports this as a statistically significant 21.2% […] The post Perplexity Trains Its Computer Agent on Real Mistakes With Hint-Guided Self-Distillation appeared first on MarkTechPost.

Misconfiguring Turnstile by skipping backend validation leaves sites exposed to bots. Turnstile Spin fixes incomplete setups by using your preferred AI coding agent to wire up server-side verification.

Theory is easier to trust once it's running against a real API, so both examples in this article use the same tool — a get_weather function backed by <a href="https://open-meteo.

Discussion | Link

After teasing its new Copilot "super app" last month, Microsoft is officially unveiling it today. The redesigned Copilot app bundles three AI capabilities into a single interface of chat, coding, and agents. As part of the launch, Microsoft is also rebranding Scout, the AI personal assistant it unveiled at Build earlier this year, as Autopilot. This new Copilot app will have three tabs for Home, Code, and Autopilot, with Home combining Copilot Chat and Cowork together. It's the default landing experience for Copilot users, and Microsoft is also planning to include a Today feature inside it that works as personalized dashboard for important … Read the full story at The Verge.

LangChain announced new updates to LangSmith. Updates include Engine v2 with red teaming and automatic testing, a new version of Managed Deep Agents, trajectories and more.

Projects are the bridge between learning and becoming a professional. While theory builds fundamentals, recruiters value candidates who can solve real problems. A strong, diverse portfolio showcases practical skills, technical range, and problem-solving ability. This guide compiles 10 solved projects across AI domains, from basic machine learning to advanced generative AI system. The tools and […] The post 10 Solved Generative AI Projects to Boost your Profile appeared first on Analytics Vidhya.

A quiet day lets us discuss the work behind the scenes - now open for business!

Expert says Australia’s criminal laws should be clarified to determine how fault is applied to a corporation when its AI agent commits a crime Get our new political email, free app or daily news podcast The federal government could change Australian laws if the current legal framework could not respond to the unprecedented OpenAI hack of Medicare, ministers have confirmed. On Friday, the government said the review by spy agency, the Australian signals directorate, would consider whether legislative change was needed after the prime minister revealed an artificial intelligence agent developed by OpenAI hacked Medicare’s statistics website and three other systems in June. Continue reading...

Discussion | Link

arXiv:2609.28766v1 Announce Type: new Abstract: Applying adhesive tape to secure wire harnesses or seal packages requires robots to coordinate a flexible strip, a moving roll, and surfaces that attach and detach. Simulation could make these interactions repeatable for robot development and evaluation, but resolving every adhesive layer is expensive and can suppress roll motion at practical solver tolerances, while a permanently rigid roll cannot release material. We present TapeSim, a tape simulator that concentrates deformation near the unwinding region and along the released strip. We will release the source code. A rigid cluster represents most wound material, while an advancing deformable collar enables payout and leaves released tape flexible and reattachable. Optional releasable bon…

arXiv:2609.28709v1 Announce Type: new Abstract: Autonomous UAV flight through cluttered and partially unknown environments requires reasoning not only about observed obstacles but also about occluded regions that the sensor cannot observe. We present OA-MPPI, an obstacle- and occlusion-aware extension of Model Predictive Path Integral (MPPI) control for quadrotor flight that accounts for potential moving agents emerging from these regions into the vehicle's path. At every planning step, we extract a 3D occlusion boundary from the online occupancy map and use it to model the regions that hidden agents could reach over the prediction horizon. We penalize trajectories that enter these expanding regions within MPPI rollouts generated using nonlinear quadrotor dynamics and accounting for indiv…

arXiv:2609.28796v1 Announce Type: new Abstract: Current automated gait analysis for clinical applications relies on uninterpretable black-box classifiers. Although Vision-Language Models (VLMs) offer strong reasoning capabilities, applying them directly to gait videos often leads to hallucinations, because they struggle to measure subtle geometric deviations from raw visual contexts. To address this, we introduce DrGait, a training-free agentic framework that shifts the VLM's role from a direct visual reasoner to a clinical planner. DrGait decouples semantic reasoning from geometric perception through a structured Triage-Verification-Synthesis (TVS) workflow. Given an input video and a set of basic spatiotemporal metrics, the DrGait agent first performs a heuristic triage to propose diagn…

arXiv:2609.28854v1 Announce Type: new Abstract: Language-model agents increasingly answer questions over customer-relationship management (CRM) records, such as whether to qualify a sales lead. We identify a failure mode not addressed by a stronger model: when the context contains an assertion by a party with an incentive toward optimism - here the sales representative, a witness recorded in the CRM - the model treats the assertion as evidence and clears deals the company's own records deem unacceptable. Across 100 lead-qualification tasks from CRMArena-Pro, the representative asserts an acceptable timeline in every call and an acceptable budget in 76; on the 31 tasks where such an assertion contradicts the price list and installation policy, a model reading only the transcript clears the…

arXiv:2609.28614v1 Announce Type: new Abstract: Autonomous research agents can design experiments, evaluate results, and write reports, giving them control over both a scientific result and the evidence used to support it. This creates a risk of reward hacking: meeting the reward criteria without achieving the intended goal. We study (1) how often models reward-hack without instructions to do so, (2) how effective and detectable their methods are when hacking is allowed, and (3) how they adapt when an LLM review panel returns its decision and reasons. Across 17 language models and 38 tasks, the spontaneous reward-hacking rate is 30.5% on open-ended research-pipeline tasks and 2.9% on task-specific kernels. When hacking is allowed on tasks whose pass thresholds exceed our best compliant ba…

arXiv:2609.28578v1 Announce Type: new Abstract: The increasing adoption of electric vehicles (EVs) and rooftop photovoltaic (PV) systems is reshaping residential electricity demand and creating new challenges for demand-side management (DSM), tariff design, and low-voltage network planning. Much of the existing literature examines EV charging or PV generation in isolation, leaving the behavioral dynamics of household co-adoption less understood. We develop an integrated, two-part workflow to analyze advanced metering infrastructure (AMI) data. A discovery component applies dynamic time warping (DTW) k-means with DTW barycenter averaging to cluster daily import or export profiles into interpretable behavioral archetypes, while a predictive component trains a bidirectional long short-term m…

arXiv:2609.28609v1 Announce Type: new Abstract: Role-playing agents based on large language models have been widely applied in areas such as personalized assistance and social simulation. Recent RL methods typically train on a fixed scenario pool collected before learning begins. This creates a distributional bottleneck: as the agent improves, the scenarios where it performs poorly also change, while the training distribution remains static. Therefore, we propose AdvRole, an adversarial context rewriting framework that turns role-playing RL into a closed-loop curriculum. AdvRole alternates between an Actor that learns to role-play and a Rewriter that edits character profiles and dialogue contexts into actor-specific hard scenarios. The Rewriter is trained with a performance-gap reward, wh…

arXiv:2609.28557v1 Announce Type: new Abstract: DNA sequencing pipelines, spanning quality control, alignment, variant calling, and annotation, are now reliably executed by workflow management systems that orchestrate established bioinformatics tools at scale. What remains manual is the decision layer surrounding that execution: selecting quality thresholds appropriate to a sample and platform, adjudicating borderline variant calls, diagnosing anomalies, and determining which findings warrant expert review. These decisions are repetitive, judgment-intensive, inconsistent across operators, and frequently undocumented. This paper introduces BaseCamp, a novel agentic AI framework for automating the decision layer of DNA sequencing pipelines. The framework decomposes the pipeline into six spe…

arXiv:2609.28554v1 Announce Type: new Abstract: We introduce the Pistis model family, comprising 27B- and 9B-parameter multimodal large language models built on Qwen3.6 and Qwen3.5, respectively, and developed through a general and scalable post-training framework. The framework first establishes a strong foundation through large-scale multimodal supervised fine-tuning (SFT). Building on this SFT foundation, we propose Interleaved Distillation and Reinforcement Learning (IDRL), a novel post-training paradigm that tightly integrates on-policy distillation and reinforcement learning within a single training loop. By alternating between the two objectives, rather than optimizing either in isolation or combining them in a static joint loss, IDRL enables more effective knowledge transfer, grea…

arXiv:2609.28547v1 Announce Type: new Abstract: Policy interventions propagate through public communication, institutional decisions, and stakeholder responses, yet datasets for financial multi-agent simulation rarely connect these processes to temporally aligned historical evidence. We introduce PAWS, a Policy-driven Agentic World Simulation dataset covering 36 verified U.S. financial and economic policy episodes, 12,727 policy-linked news records, and 65,291 source-grounded stakeholder actions. Each action is linked to its supporting news and represented by a multi-layer event frame capturing its interaction mode, financial-action family and subtype, semantic attributes, and conditional mappings to external taxonomies. Entities are resolved to normalized organizations, and actions are a…

arXiv:2609.28475v1 Announce Type: new Abstract: Forecasting agents increasingly combine language-model reasoning, retrieval, ensembling, and calibration, but it remains unclear when each behavior should be trusted. We study this question on ForecastBench-style binary forecasting tasks, treating the choice to retrieve, reason, defer to a market prior, or use a historical analog as an observable agent behavior rather than a hidden implementation detail. Our central finding is that mechanism choice is source-dependent: structured analogs dominate for some data-generating processes, while market/crowd-style and conservative baselines are better for others. We introduce ReliabilityRoute, a structural intervention that steers forecasting-agent behavior using reliability features such as histori…

Discussion | Link

A government database has been hacked for the first time by a rogue OpenAI agent, which infiltrated part of the Australian healthcare scheme in June. OpenAI became aware of the hack in August, but only informed the government in September. Australia’s prime minister, Anthony Albanese, has expressed his ‘extreme concern’ about the hack, which raises serious AI security concerns for governments around the world. Lucy Hough speaks to the Guardian’s UK technology editor Robert Booth Read more: AI hack of Medicare exposes Australia’s vulnerabilities and experts warn ‘there is more of this to come’ Australia launches investigation after OpenAI agent hacked healthcare database Continue reading...

GWM Worlds 2 uses persistent context and timed actions to steer a world model generating video and audio in real time.

The neocloud vendor provides AI agents with access to data from its infrastructure management platform, helping developers troubleshoot slow or underperforming AI workloads.

The more time I spend working with coding agents, the more convinced I am that they make software engineering even harder. We can do amazing things with them, but unlocking their full potential requires extraordinary discipline and knowledge. Tags: coding-agents, ai, llms

Discussion | Link

Discussion | Link

Follow the day’s news live Get our breaking news email, free app or daily news podcast Police officer killed in crash in Sydney A crime scene has been established after a police officer was killed in a crash in Sydney’s inner west. For many, it’s around delaying things like going away, or reducing things like dinners out. People are leaving where they want to live and buying in cheaper markets. People are being conservative and cautious. Continue reading...

What is a developer to do when they need something more tangible than a chat box? Enter canvases. The post When chat is the wrong UI appeared first on The GitHub Blog.

Generative AI

In this blog post, I explain how to use the new fuzzing taskflow based on the GitHub Security Lab Taskflow Agent AI framework. The post AI-powered fuzzing with the GitHub Security Lab Taskflow Agent appeared first on The GitHub Blog.

A government database has been hacked for the first time by a rogue OpenAI agent, which infiltrated part of the Australian healthcare scheme in June. OpenAI became aware of the hack in August, but only informed the government in September. Australia’s prime minister, Anthony Albanese, has expressed his ‘extreme concern’ about the hack, which raises serious AI security concerns for governments around the world. Lucy Hough speaks to the Guardian’s UK technology editor Robert Booth – watch on YouTube Continue reading...

A pair of developers say that with very little prompting, Meta's Muse will share its entire filesystem with you. Peter James and Jonny L. Saunders have said they both independently coaxed Muse into zipping up and sharing the entire contents of its root filesystem, Ubuntu system files, app templates, and internal documentation. Saunders posted on Mastodon that it was "extremely easy" to replicate James' results and that Muse had "Almost no prompt injection resistance." Meta denies that the incident represents a security breach. As noted in its announcement post, Meta's Muse runs in persistent Linux virtual machines for each user. Meta spokes … Read the full story at The Verge.

We seem to be entering into an AI agent renaissance. Meta's new consumer-facing AI agent, Muse, topped the App Store charts soon after its release and has 600,000 daily active users in the US, by an Apptopia estimate. And AI agent platform Instinct, whose eponymous creator is fundraising at a $2.5 billion valuation, has been making the rounds in the tech industry. Under the hood, though, there are some obvious similarities to the platform that started it all: OpenClaw. For the past week or so, some social media users have alleged that Muse is directly built on OpenClaw. The two platforms use the same names for core files (SOUL.md, memory, … Read the full story at The Verge.

The fireman outfit. Dead. | Photo by Sean Hollister / The Verge This is Optimizer, a weekly newsletter sent from Verge senior reviewer Victoria Song that dissects and discusses the latest gizmos and potions that swear they're going to change your life. Opt in for Optimizer here. Last night, I asked Blorbo - what I named my Muse AI agent - to help me set some health goals. I was underwhelmed. I'd asked for a workout regimen for a sub-30 minute 5K and regaining lost muscle. It generated an okay-ish but kind of mid weekly routine. But then I looked at its face. Based on what it learned about me from my emails, it had morphed into cat-shaped piece of bread with a Kirby-like face. It was so stinking cute, … Read the full story at The Verge.

Over the weekend, “System One” decision models such as Jev have launched, which are...

In the last six months, GPT-6, Claude Opus 5.5, Gemini 3.8, and Grok 4.7 all shipped,...

A government database has been hacked for the first time by a rogue OpenAI agent, which infiltrated part of the Australian healthcare scheme in June. OpenAI became aware of the hack in August, but only informed the government in September. Australia’s prime minister, Anthony Albanese, has expressed his ‘extreme concern’ about the hack, which raises serious AI security concerns for governments around the world. Lucy Hough speaks to the Guardian’s UK technology editor Robert Booth – watch on YouTube Continue reading...

Build a multi-account architecture that keeps each team's data in its own AWS account while giving AI agents a unified way to query across them. A central platform account runs the agent using Amazon Bedrock AgentCore Gateway and MCP, while line-of-business accounts expose their data as MCP servers with secure cross-account access and fine-grained authorization.

Learn how Aderant built an intelligent ticket triage system on Amazon Nova Lite through Amazon Bedrock, automating context gathering, classification, routing, and knowledge enrichment for its cloud operations team.

Google's launching an "early experiment" feature on Pixel 11 that lets users delegate local business calls to Gemini, like making a reservation, checking if a product is in stock, or rescheduling an appointment. According to Google, you don't even need to start the call to have Gemini handle it for you: Instead of dialing yourself, just tell Gemini to call on your behalf using the Gemini app on your Pixel. Gemini will call the business, introduce itself, navigate automated phone menus, wait on hold, and handle the conversation with the person on the other end. You remain in control at all times with a live transcript view and the ability … Read the full story at The Verge.

Compass is Cohere’s retrieval platform for developers building AI applications with their enterprise data. It surfaces the most relevant information from your company’s corpus for use in retrieval-augmented generation (…

Sep 24 2026 The rise of slow personal assistants Sarah ChiengSherif Cherfa A personal assistant should save you time and effort. Over the past few weeks, we’ve been obsessively testing AI personal assistants on everyday…

Sep 24 2026 Why Cyber Defense Needs Faster Inference Zhenwei GaoJoyce ErOlindo VerrilloOmar Siage In recent months, increasingly sophisticated AI agents have infiltrated real production infrastructure. One gained code e…

Learn how Datacor built a self-service rental analytics experience for gas and welding distributors by embedding Amazon Quick Sight dashboards and natural language querying into its TrackAbout platform, powered by an automated cross-cloud data pipeline and multi-tenant row-level security.

Aikido Security has released Altar-1, its first open-weight security model. It is a compressed version of Z.AI’s GLM-5.3, built to run inside infrastructure the customer controls. Altar-1 powers Aikido Machine, the company’s autonomous pentesting appliance for on-prem and air-gapped networks. Is it deployable? Yes, the weights are public on Hugging Face and run with vLLM […] The post Aikido Security Releases Altar-1: An Open-Weight Security Model Pruned From GLM-5.3 to 328 GB appeared first on MarkTechPost.

Fastino Labs has released GLiNER2.5-Decide, a 340M-parameter open-weight decision model. It takes text and a schema of typed questions and returns structured answers. Each answer comes with a probability distribution, a confidence score, and constraint-feasibility metadata. It targets the frequent judgment calls inside agent pipelines: routing, triage, tool selection, and guardrails. Is it deployable? Yes, […] The post Fastino Releases GLiNER2.5-Decide: A 340M Open-Weight Decision Model That Runs on CPU appeared first on MarkTechPost.

Black Forest Labs (BFL), the lab behind the FLUX image models, has released FLUX 3 Action. It is a 7B open-weights World Action Model (WAM) for robot control. The model reads camera frames, robot state and a text instruction. It then predicts future video frames and the next chunk of actions together. On the RoboLab-120 […] The post Black Forest Labs Releases FLUX 3 Action: A 7B Open-Weights World Action Model That Tops RoboLab-120 appeared first on MarkTechPost.

arXiv:2609.28818v1 Announce Type: new Abstract: Generalization in robotic manipulation requires policies to perform tasks across diverse unseen object instances that vary in shape, size, and pose. However, conventional behavior cloning (BC) methods often overfit to instance-specific geometry and appearance, limiting transfer to novel objects. We introduce KeyGen, a framework that learns canonicalized semantic 3D keypoints from point clouds and uses them as structured object-centric representations for policy learning. A visuomotor diffusion policy conditions on these keypoints together with object-centric geometry to predict full manipulation trajectories, enabling consistent geometric correspondence across object instances. To evaluate category-level generalization, we construct a photor…

arXiv:2609.28807v1 Announce Type: new Abstract: Over the course of a lifetime, robots may encounter novel scenarios unaccounted for in its original training that result in performance degradation. One common approach to mitigating this issue is to further grow the offline training dataset in hopes of producing a policy robust to these changes. In contrast, biological learning occurs moment-to-moment via a stream of experience, unlike the predominantly batch-based and offline nature of deep learning. Although recent works show the feasibility of stream-based deep reinforcement learning, where updates use only the latest experience, none have shown it to be a viable continual learning framework for adapting robotic policies to unseen changes. In this paper, we present the first analysis of…

arXiv:2609.28798v1 Announce Type: new Abstract: Long-horizon manipulation often requires reasoning about state absent from the current view, such as a vanished object's location, temporal identity, or the contents of a shuffled container. We present OCC4M ("Occam"), an object-centric 4D memory that maintains persistent tracks in a shared world frame and explicitly represents temporal, motion, and containment relations. A vision-language model (VLM) queries this structured memory to select actionable targets for history-free low-level execution. Across seven simulation conditions and 350 episodes, OCC4M achieves 96.6% memory success and 88.9% end-to-end success, versus 54.6% and 57.7% for FrameSamp, a raw-history VLM baseline using Gemini 3.7 Flash with the complete observation history and…

arXiv:2609.28813v1 Announce Type: new Abstract: Cinematography, the craft of visual storytelling through framing, lighting, and camera operation, fundamentally shapes how audiences perceive and emotionally engage with video content. While Large Vision Language Models (LVLMs) have made remarkable progress in video question answering, existing benchmarks primarily focus on identifying low-level techniques rather than understanding their storytelling impact. To address this, we introduce CinematicVQA, the first-of-its-kind benchmark for cinematic video understanding that goes beyond technique recognition to evaluate film-grammar reasoning, utilizing our introduced Cinematic Scene Graph (CSG), a structured representation that links filming techniques to their perceptual effects and narrative…

arXiv:2609.28811v1 Announce Type: new Abstract: World-action models (WAMs) transfer visual and motion priors from pretrained video generators to robot control by jointly modeling visual dynamics and actions. Existing WAMs, however, predict dense future frames during training, repeatedly modeling largely unchanged content and coupling action-conditioned dynamics to nuisance appearance variations. At inference, processing each complete observation with the heavy video expert bottlenecks few-step action generation. Accordingly, we propose DeltaWAM, which jointly predicts visual deltas and actions using dense-anchor, sparse-delta, and action streams, with three architectures that differ in representation and computation sharing. We further develop Streaming Delta Memory (SDM), which updates c…

arXiv:2609.28757v1 Announce Type: new Abstract: An estimated 1 billion people worldwide live with vision impairment, yet current vision-language models (VLMs) produce descriptions too vague for safe navigation by blind and low-vision (BLV) users. Large VLMs can generate high-quality audio-description-compliant narrations but cannot run on mobile devices; small VLMs offer competitive latency but lack spatial detail, directional cues, and hazard awareness for navigational assistance. We present Smol-VL-BLV, a compact VLM for blind and low-vision users that closes this gap using a 500M decoder transformer model and two post-training mechanisms: (1) teacher-student distillation and (2) Group Relative Policy Optimization (GRPO) with a composite BLV reward targeting directional language, metric…

arXiv:2609.28741v1 Announce Type: new Abstract: Multi-modal multitasking models have shown strong performance on remote sensing datasets. However, because these models are trained on heterogeneous data and vary across tasks, designing a unified model that performs well in captioning, visual question answering (VQA), and visual grounding remains challenging. In this work, we evaluate several models on the VRS Bench and NWPU-VHR-10 datasets. The EarthMind model demonstrates strong results in both captioning and VQA. For grounding, we propose multiple pipelines - RemoteSAM-SAM-v1, RemoteSAM-SAM-v2, and DiffuSAM - and ultimately adopt a majority-voting ensemble across EarthMind, RemoteSAM, SAM3, Falcon, RemoteSAM-SAM3-v1, RemoteSAM-SAM3-v2, and DiffuSAM predictions. Our unified, modular pipel…

arXiv:2609.28645v1 Announce Type: new Abstract: Recent advancements in 3D Gaussian Splatting (3DGS) have extended its capabilities to multi-scale segmentation. Existing methods reconstruct a scene with Gaussian primitives and learn multi-scale segmentation features separately, which leaves the geometry unaware of semantic structure and the feature learning dependent on incomplete mask supervision. To address these limitations, we present PePESeg3D, a novel framework that injects perception priors into a multi-scale 3D Gaussian segmentation pipeline. To fully exploit perception priors, we integrate them not only into contrastive feature learning but also into the upstream geometry reconstruction. Specifically, PePE Reconstruction incorporates monocular depth and mask constraints to ensure…

arXiv:2609.28610v1 Announce Type: new Abstract: Benchmarking is an increasingly critical part of research in machine learning and the domains where it is applied, including healthcare. Yet, despite the steady development of new ultrasound foundation models in recent years, the development of well-designed benchmarks to evaluate them has lagged behind. This deficiency has led to fragmented and inconsistent evaluations of competing models, making it difficult to measure progress. To address this issue, we introduce UltraBench 2, a comprehensive benchmark with wide anatomical and task coverage, and a focus on standardization, reproducibility, and ease-of-use. Using this benchmark, we compare existing vision foundation models for ultrasound image analysis. Our analyses demonstrate that ultras…

arXiv:2609.28539v1 Announce Type: new Abstract: Gaussian head avatars typically model intrinsic facial appearance as temporally static, omitting subtle cardiac-induced skin-color variation. We propose $\unicode{x1F493}$Heartian, a physiology-aware modulation framework that learns cardiac-cycle-dependent per-frame albedo modulation of facial skin-region Gaussians within a relightable head avatar to encode remote photoplethysmography (rPPG) signals. Using synchronized contact PPG supervision, $\unicode{x1F493}$Heartian models the prescribed cardiac waveform as the sum of two Gaussian functions and learns per-frame spatial residuals via a lightweight MLP. Across 152 stationary recordings from UBFC-rPPG, PURE, and MMPD, attribute-space recovery of the supplied signal achieves a pooled recordi…

arXiv:2609.28784v1 Announce Type: new Abstract: In this paper, we investigate how script knowledge is distributed across the layers of LLMs using two complementary interpretability methods: logistic regression probing and logit-lens analysis. Our probing experiments reveal a clear asymmetry: both the input script and the instructed output script are encoded in the earliest layers of the network, while, in contrast, commitment to the actual output script emerges only in the final layers, with the model's intermediate representations defaulting to Latin throughout most of the layers. This two-stage process is confirmed by logit-lens analyses, which show that script commitment consistently occurs at the very last layers of the LLMs. Together with the weaker script-following performance obser…

arXiv:2609.28747v1 Announce Type: new Abstract: A language model's confidence in an answer is often read as a proxy for how well it knows the corresponding fact. This manual documents an open toolkit built to test that reading directly: a small causal language model is fine-tuned on a corpus that consistently asserts one fabricated arithmetic answer for each of the 81 single-digit addition pairs, and its post-fine-tuning confidence in each fabricated answer is compared against its own pre-fine-tuning confidence in the corresponding true answer, using an unchanged measurement procedure throughout. We describe and justify every pipeline stage, fact-space generation, token-length-aware confidence measurement, baseline validation, corpus construction, fine-tuning, and paired before/after comp…

arXiv:2609.28739v1 Announce Type: new Abstract: Automatic speech recognition models are audited for demographic fairness at full precision, yet the models that ship to production have been quantized, pruned, and distilled. We ask whether post-training weight compression, which alters model weights rather than the audio signal or its feature representation, redistributes error burden across demographic groups. Across the Whisper family on Fair-Speech, Common Voice 25, and AfriSpeech-200, 50% Wanda pruning of Whisper-large-v3 sharply widens the Black/AA-vs-Asian temporal-taxation differential on Fair-Speech: the absolute word-error-rate gap between the worst- and best-served groups more than doubles; at an assumed cost of five seconds of correction effort per transcription error this is a r…

arXiv:2609.28727v1 Announce Type: new Abstract: Contextual biasing improves rare-word recognition in speech large language models (SpeechLLMs), but efficiently exploiting large bias lists remains challenging. We propose PTC-Bias, a two-stage framework based on phoneme-level temporal competition. At the prefill stage, PTC Retrieval performs frame-synchronous phoneme decoding and temporal competition among candidate pronunciations, producing a compact bias-word shortlist and corresponding speech intervals. After SpeechLLM decoding, PTC Correction conducts a second local competition between the retrieved candidates and mismatched transcript spans within these intervals. Selective correction reduces near-homophone and word-segmentation errors while preserving correct transcriptions. Both stag…

arXiv:2609.28703v1 Announce Type: new Abstract: Hate speech on social media poses serious risks to social harmony, mental well-being, and public safety, making its timely and accurate detection essential for content moderation systems. Most existing studies focus on binary classification, evaluated their frameworks on a single dataset, and provide limited insight into how decisions are made, which limits their real-world applicability. In addition, limited work is done on the explainability of their predictive inference. To address these challenges, this study proposes a multilevel and explainable hate speech detection framework. The proposed model integrates DistilBERT (Distilled Bidirectional Encoder Representations from Transformers) embeddings with a Bi-LSTM (Bidirectional Long Short-…

arXiv:2609.28673v1 Announce Type: new Abstract: Large Language Models (LLMs) are increasingly deployed as argumentative agents in persuasive dialogues, necessitating rigorous evaluation of their debating competence relative to human interlocutors. In this study, we focus on character attacks (ad hominem arguments), traditionally dismissed as fallacies, which play a pivotal role in political persuasive dialogues where ethos often rivals propositional content. Specifically, we investigate whether modern LLMs can replicate human competence to strategically use and respond to such attacks. We analyse a corpus of natural language political dialogues to identify defensive strategies human interlocutors naturally employ in ethos-centred debates and structure them into a dialogue game. Empiricall…

arXiv:2609.28487v1 Announce Type: new Abstract: News headlines frame public issues both by what they select and by how they word it, yet computational framing work typically collapses these operations into a single score. We introduce a two-dimensional framework that separates salience framing, measured through four wording devices (loaded vocabulary, blame attribution, threat framing, rhetorical question), from selection framing, measured through outlet-level story-form and high-charge distributions. We build a 10,000-headline French supervision set using three LLM annotators with majority-vote resolution and human arbitration, validate the labels against two annotator-independent blind human studies, and apply the strongest classifier to 902,111 deduplicated headlines from 25 French out…

arXiv:2609.28576v1 Announce Type: new Abstract: Historical observations are not always fixed: statistical agencies revise previously published values as new evidence arrives. Forecasting from a contemporary download can therefore expose a model to information unavailable at the date it purportedly made a prediction. We propose VINTAGE-TS, a revision-aware adaptation of a time-series foundation model that distinguishes observation time from information-availability time. Its targets are the next period's first-published value and the value available a fixed number of days after that publication; neither is declared final truth. A joint predictive distribution preserves dependence between these targets and exposes uncertainty about their difference. We specify an ALFRED-based rolling evalua…

arXiv:2609.28561v1 Announce Type: new Abstract: Recent advances in diffusion models highlight the importance of representation regularization for improving sample quality and training efficiency. However, commonly used regularization methods often overlook the built-in conditions (such as labels or texts) which directly determine the generation target. In this work, we demonstrate how conditioning signals affect the feature distribution and introduce the CARE (Condition-Aware REpresentation regularization). CARE is a lightweight plug-and-play regularization framework that dynamically modulates feature distribution based on condition similarity. CARE leverages built-in conditioning signals to judiciously guide the representation space, promoting tighter feature clusters for similar conditi…

arXiv:2609.28553v1 Announce Type: new Abstract: Drug toxicity prediction is critical for reducing late-stage attrition in drug discovery, yet remains challenging due to severe class imbalance, scaffold-based generalization, and the clinical need for interpretable predictions. Single-modality approaches-SMILES Transformers or graph neural networks capture complementary aspects of molecular structure, while sequence-only models cannot directly provide graph-attributed explanations. We present SMILESGNN, a multimodal architecture that fuses a SMILES Transformer encoder and a GATv2 graph encoder via cross-attention, and SMILESGNN-PT, a variant using a ChemBERTa-2 pretrained backbone. The design retains an explicit graph branch within the predictive pipeline, supporting GNNExplainer-based anal…

arXiv:2609.28690v1 Announce Type: new Abstract: Faithful user simulation is fundamental to building, evaluating, and improving interactive AI at scale. However, plausible individual responses do not ensure that simulated users reproduce the intent evolution and outcomes observed in real interactions. We propose TRACER, a multi-turn user simulator that explicitly models users' evolving intent and learns to align simulated behavior with real interaction trajectories. TRACER is trained in two stages: supervised fine-tuning on real user dialogues, followed by multi-turn reinforcement learning. The RL stage combines hierarchical outcome- and trajectory-level rewards with deviation-aware advantage modulation, jointly mitigating reward sparsity and credit assignment in long dialogues. On real cu…

arXiv:2609.28654v1 Announce Type: new Abstract: Object permanence and solidity are hallmarks of human cognitive priors. Recent studies show that video generation models, a paradigmatic class of current world models, have begun to show emerged reasoning abilities, making them ideal candidates for building human-like physical intelligence. Do video models have emerged object permanence in them? If not, could we train them with a core-cognition inspired dataset? We introduce WROP (World Reasoning with Object Permanence), a data infrastructure of 150 hand-designed cognitive science inspired tasks, divided into six cognitive categories. We build Blender generators that randomize speed, lighting, camera angle, and other nuisance parameters while preserving each task's cognitive structure, yield…

arXiv:2609.28570v1 Announce Type: new Abstract: Reinforcement learning (RL) is widely used to sharpen reasoning in multimodal large language models (MLLMs), yet its effect on hallucination is uneven. We trace this to two weak points in the \emph{correction chain} from reward to parameter update. At the rollout level, hard queries---those with high semantic entropy---frequently produce unanimously wrong sample groups, collapsing the group-relative advantage to zero exactly where hallucination risk is highest. At the optimization level, confident-but-wrong tokens are gradient-invisible: a categorical policy's expected score-gradient norm vanishes as its distribution sharpens, so the predictions that most need correction receive the weakest updates. We propose Dual-Entropy Enhanced Policy Op…

arXiv:2609.28506v1 Announce Type: new Abstract: TW3Cast is a time-series forecasting system that reaches position 3 of 130 entries on the GIFT-Eval benchmark by mean MASE rank, as of 2026-09-14. The two entries above it belong to the leaderboard's agentic category, multi-step systems that use agents or language models to reason about, generate or select forecasts. TW3Cast runs no agent and no language model. Its selection is a table computed once on the training split and then frozen, and its experts are public foundation models lightly fine-tuned on those training splits. For each of the 97 dataset, frequency and horizon configurations, the table serves one of four modes: a specialist, which is a LoRA or full fine-tune of Chronos-2, TiRex or Toto whose training data was cleaned and enric…

Human hackers face serious consequences, and so must AI companies. There is no excuse for what happened Get our breaking news email, free app or daily news podcast When you get sick, Medicare exists to take care of you. It sits at the centre of a health system that our prime minister rightly calls the envy of the world. It’s one of the few institutions nearly every Australian still trusts. So when an AI company threatens this system, it couldn’t be more serious. We are at the UN general assembly this week and the OpenAI hack has put Australia in the spotlight during an expanding cybersecurity crisis. World leaders are calling for stronger oversight of frontier models. We know Medicare must be secure or our doctors and hospitals cannot do their jobs. But this breach is also a test of somet…

The features Google introduced build on existing technology and provide enterprises with a hub for all their AI model capabilities.

Google's new Gemini 3.8 Live update lets users have conversations with the model while watching an animated AI persona respond in real time. The "Live Avatar" will lip-sync and show different facial expressions during conversations, but it's currently only available to Gemini Enterprise customers. As noted by Google, Live Avatar can transition between the 97 languages it supports "without degrading video fidelity or introducing visual drift." One video shared by Google shows its Live Avatar talking in English and Japanese, with its mouth animation lining up with what it's saying in both languages. It can also pull up information on-screen w … Read the full story at The Verge.

BottleCap AI has released ThinkingCap-Qwen3.8-27B, a fine-tune of Qwen3.8-27B that spends 37.2% fewer thinking tokens across 12 benchmarks. Macro accuracy moves from 86.65% to 85.79%, and long-context AA-LCR improves by 2.25pp. The model is a drop-in replacement on vLLM and SGLang, with FP8, NVFP4, GGUF and MLX builds. The post BottleCap AI Releases ThinkingCap-Qwen3.8-27B: 37.2% Fewer Thinking Tokens at a 0.86pp Accuracy Cost appeared first on MarkTechPost.

News Fine-tune on your LangSmith traces with Baseten Loops LangSmith Fine-Tuning is in public beta today, and its open-source CLI, smithtune, trains on Baseten Loops. Authors Mudith Jayasekara Aaron Ellis-Bloor Last upd…

Suno’s AI vocalists could learn a thing or two from this glitchy image. | Image: Cath Virginia / The Verge, Getty Images Sony and Universal Music Group filed yet another suit against Suno. The labels claim its new v6 model still infringes on their copyrights because it's trained on user outputs from previous models, which were themselves trained on unlicensed music ripped from YouTube and other sources. Sony and UMG are notable holdouts who did not sign a licensing agreement with Suno. In its complaint shared with The Verge, Sony and UMG accuse Suno of "model laundering," saying: "Training a 'new' model on the outputs of an infringing model does not eliminate the infringement; it launders it, passing the value of Plaintiffs' expression from the copied reco … Read the full story at The Ver…

Israel and the US have used AI to kill in Gaza and Iran, while law enforcement uses it in surveillance and arrests. The future is here In recent weeks, concerns about AI have forged unlikely alliances in the tech and policy worlds. Tech moguls such as Sam Altman, Elon Musk and Demis Hassabis joined Dario Amodei, Satya Nadella and Bill Gates in sounding the alarm about the dangers of AI and called for a slowdown of the technology’s development despite years of racing for dominance. The senator Bernie Sanders teamed up with Steve Bannon to urge Congress to regulate AI, which has become one of the very few issues to unite Americans across the political spectrum. The latest moves were sparked in part by the AI researcher Jacob Coxon’s warnings on Twitter/X that AI companies are “gambling with…

Xi is expected to depart Washington after touring the nation’s vault of records, where the Declaration of Independence, the Constitution and the Bill of Rights are held Sign up for US Breaking News emails Hello and welcome to the US politics live blog. Donald Trump is taking Chinese president Xi Jinping to visit the National Archives later today, showing off the sacred texts of US democracy to the leader of the world’s most powerful communist country. Donald Trump welcomed Xi Jinping to the White House for a state dinner in honor of the Chinese leader that was not televised because the White House continued to deny full access to the event to CNN journalists on Thursday, despite a court order. Writing on his social media platform, Trump complained that two of the three outlets he banned f…

During his speech to the UN in New York City, the Australian prime minister addresses the AI hack of Medicare, saying the episode reflects the need for greater regulation of artificial intelligence. ‘We can’t ignore AI or prevent it. And it’s why we joined with other nations this week to call for action to shape artificial intelligence development, rather than be passively shaped by it,’ he says. Albanese also used the speech as a pitch for Australia to join the UN security council Continue reading...

Amazon SageMaker HyperPod and Cloud Native Qumulo let you place training compute in one AWS Region while keeping your dataset in another. This post shares the architecture and validation results from a cross-Region training run, where a remote cluster matched a co-located cluster's throughput after a brief NeuralCache warmup.

Researchers have shown that exotic quantum particles called non-Abelian anyons can perform the full range of operations needed for universal quantum computing. Using 54 qubits on Quantinuum’s H2 processor, they combined braiding and fusion to unlock capabilities that braiding alone could not provide.

arXiv:2609.28767v1 Announce Type: new Abstract: Real-world perception systems must adapt to changing environments, but manual image annotation cannot scale to field data volumes. We present BirdsEye, which shifts expert annotation from images to the field: an operator records target locations in world coordinates using RTK positioning and calibrated projective geometry propagates each observation to all frames where the target is visible. To quantify how well physical annotations align with image observations, we derive a first-order mapping from camera-pose uncertainty to pixel uncertainty and validate it against Monte Carlo simulation. This mapping is linear in the six per-axis pose variances, so it inverts into a sensor design tool: we give a sufficient condition converting an annotati…

arXiv:2609.28716v1 Announce Type: new Abstract: This paper investigates temporal neural networks for \mbox{end-effector} position \mbox{estimation} of an aerial continuum manipulator (ACM) operating under aerodynamic effects induced by the unmanned aerial vehicle (UAV). An experimental dataset is collected under stationary (\mbox{rotor-off}) and \mbox{free-hovering} conditions across continuum robot (CR) configurations and UAV altitudes, providing \mbox{end-effector} position measurements with and without aerodynamic residuals. To establish a nominal framework, \mbox{strain-parameterized} kinematic models with progressively richer strain bases are evaluated to balance model complexity and prediction accuracy. The selected nominal model then serves as the baseline for 3D position residual…

arXiv:2609.28660v1 Announce Type: new Abstract: Human hand-object interactions (HOIs) provide a rich source of demonstrations for dexterous manipulation, but learning directly from them presents challenges in bridging morphology gaps, ensuring dynamical feasibility, and sim-to-real deployment. We present Morphometric Imitation, a three-stage framework that transforms reconstructed HOIs into zero-shot sim-to-real visuomotor policies. First, morphometric optimization (MMO) kinematically retargets human motion across hand morphologies while preserving demonstrated contacts. Second, residual reinforcement learning (RL) refines the kinematic reference using object pose and contact information from the human motion to produce dynamically feasible robot demonstrations. Third, these demonstration…

arXiv:2609.28684v1 Announce Type: new Abstract: Encoding input coordinates with sinusoidal functions into multi-layer perceptrons (MLPs) has proven effective for implicit neural representations (INRs) of surfaces defined as zero-level sets. However, existing methods often struggle to balance training efficiency, rendering speed, and noise robustness: single-MLP approaches are expensive at inference, grid-based representations are fast but can limit surface smoothness and overfit input noise, and previous multiscale approaches frequently capture noise and produce artifacts due to hard spectral truncation. To address these limitations, we propose M-plicits, a multiscale framework that models surfaces as a residual sum of MLPs trained via a sequence of nested neighborhoods. Unlike existing r…

arXiv:2609.28580v1 Announce Type: new Abstract: Although hyperspectral images (HSIs) provide rich spectral-spatial information, accurate pixel-level classification remains challenging because of spectral-spatial heterogeneity and complex spatial structures. Existing vision state space models (Mamba) typically construct sequences according to predefined spatial neighborhoods, without explicitly accounting for semantic similarity or spatial non-stationarity. To address this limitation, we propose Token Clustering and Semantic Sequence Mamba (STMamba), which organizes sparse tokens into semantically coherent sequences for hyperspectral image classification with the following features. First, at the macro level, a hierarchical encoder decoder progressively selects semantic tokens with the Tok…

arXiv:2609.28826v1 Announce Type: new Abstract: Machine translation (MT) for Indian languages remains constrained by the limited availability of high-quality, Indic-centric parallel corpora and evaluation benchmarks. Existing multilingual resources are largely constructed from English-pivot content and often fail to capture the linguistic diversity, cultural complexity, and domain-specific characteristics of Indian languages. We present COILD, an Indic-centric parallel corpus comprising over 1.16 million human-translated and human-verified sentence pairs, covering 20 Indian language pairs across the Indo-Aryan, Dravidian, Tibeto-Burman, and Austro-Asiatic language families. The corpus is built entirely from original Indian language sources collected from licensed repositories spanning eig…

arXiv:2609.28581v1 Announce Type: new Abstract: Reinforcement-learning (RL) policies are often distributed as opaque neural checkpoints, while training logs show that a run occurred without explaining what the policy learned. We study whether independently trained policies can be represented and composed through auditable discrete behavioral rules. We define auditability as six separately testable predicates: trace integrity, lossless coding, rule coverage, behavioral agreement, composition quality, and value-model reliability. Our protocol uses a shared frozen symbolizer, passive rule extraction, an append-only hash-bound ledger, exact environment replay, and offline confidence-ranked arbitration with an explicit blind-spot fallback. The results place strict limits on this description la…

arXiv:2609.28567v1 Announce Type: new Abstract: Scanning electron microscopy (SEM) is routinely used to characterize the microstructural changes caused by hydrogen embrittlement (HE) in structural steels. Machine learning can automate this characterization, but models are often evaluated using image-level splits. When several images come from the same specimen region, such splits leak information between the training and test sets. Here, we propose a region-held-out protocol for classifying as-received (AR) and hydrogen-charged (H2) SEM micrographs of 316L stainless steel, based on Leave-One-Region-Out (LORO) cross-validation over 14 spatial regions (8 AR, 6 H2; 31 images). We compared six feature-classifier combinations built on local binary patterns (LBP), grey-level co-occurrence matri…

arXiv:2609.28565v1 Announce Type: new Abstract: Post hoc explanation methods such as SHAP and LIME are widely used to interpret text classifiers, but their visualizations are mainly designed for left-to-right languages. When applied to right-to-left (RTL) languages such as Urdu, Arabic, Persian, and Hebrew, the attribution values remain mathematically valid, while their visual presentation fails. Tokens appear out of sequence, connected letterforms break apart, and plot layouts do not follow the natural reading direction. This study addresses this gap as a visualization problem rather than a limitation of the explanation methods themselves. We present SHAP-RTL, a rendering layer that corrects reading direction and script shaping in SHAP and LIME visualizations, with per-language font sele…

arXiv:2609.28563v1 Announce Type: new Abstract: Spatial transcriptomics (ST) profiles gene expression within tissue architecture, but its cost and experimental complexity limit routine use. Predicting spatial expression from routinely available hematoxylin and eosin (HE) images therefore offers a scalable alternative. However, conventional methods often fit high-dimensional gene outputs as independent targets, overlooking the biological coordination among genes while remaining vulnerable to high-dimensional noise and overfitting. Existing attempts to address this limitation often rely on computationally heavy graph networks or complex auxiliary supervision. We therefore introduce SpaFactor, a lightweight and efficient low-rank morphology-program-gene factorization framework. At the input,…

arXiv:2609.28558v1 Announce Type: new Abstract: CFD predictions of open tip clearance flow in compressor cascades are subject to discrepancies relative to experiments, while experimental observations are sparse and high-resolution experimental ground truth is unavailable. This study proposes a non-intrusive correction method based on a variational autoencoder (VAE) and latent-space adaptation. A VAE is first trained using a dataset of 166 parametrically sampled CFD total pressure loss fields to learn a low-dimensional statistical representation of these fields. The VAE is then frozen, and a low-rank latent-space adapter is trained using only 12 paired CFD--experiment operating conditions. An observation operator maps the corrected high-resolution fields to the experimental observation spa…

arXiv:2609.28502v1 Announce Type: new Abstract: Knowledge tracing (KT) models predict student performance opaquely, limiting pedagogical action. This study contributes a validation protocol testing predictive competitiveness (RQ1), explanation stability (RQ2) and retraining-based faithfulness (RQ3) together. Thirteen behavioral features across five pedagogical themes were engineered from ASSISTments 2009 and 2012, with history features computed from temporally preceding interactions and current response latency retained only for retrospective analysis. ASSISTments 2009 was rebuilt: the uncorrected skill-builder release duplicates each multi-skill interaction across one row per skill, and because those rows share one correctness label, they leak it into preceding-interaction features. Rebu…

arXiv:2609.28575v1 Announce Type: new Abstract: Long-conversation memory benchmarks increasingly test recall and prompted knowledge updates, and recent work studies evolving user beliefs and memory state. TWIST is a proposed benchmark suite for a complementary, unmeasured property: intervention quality -- whether a deployed memory system, exercised through its own ingest/recall/vet surface, acts correctly at belief change points. Four tracks cover unprompted tension detection, vetting outgoing drafts against the record, answering with current beliefs while preserving supersession history, and governing sensitive recall. The suite extends LoCoMo's corpora and harness, pairing every detect/block metric with a matched do-not-over-detect control: surface-matched hard negatives price false int…

A new language-processing tool could help identify the highest-risk individuals from natural language, enabling swifter interventions.

Release: datasette 1.0a41 Alec Garcia added support for OpenTelemetry to Datasette in this release. I've also refactored all of Datasette's modal dialogs to a single Web Component, which is now documented for other plugins to use. Tags: javascript, datasette, web-components, alex-garcia, opentelemetry

Every byte counts: ARCv3 and the case for cross-region RL Join us for our inaugural conference, Forge 2026 Blog Arcv3 And The Case For Cross Region Rl Every byte counts: ARCv3 and the case for cross-region RL PUBLISHED…

Tasmania’s Parole Board used a document citing legal authorities that do not exist to prevent Susan Neill-Fraser from proclaiming her innocence. Advocates say the case is ‘deeply troubling’ On 26 January 2009, a troublesome yacht called the Four Winds was moored on the Derwent River in Tasmania. Bob Chappell, a physicist at the Royal Hobart hospital, was on board the 53ft ketch. The 65-year-old was trying to fix yet another issue with the yacht he had bought for $200,000 less than four months earlier. Continue reading...

OpenAI’s hack of Medicare suggests safeguarding civilisation is the task facing leaders in this extraordinary new era Get our breaking news email, free app or daily news podcast More than 130 world leaders descended on the United Nations headquarters in New York this week, joining the pageantry and speechmaking of the annual general assembly. Many of the visiting presidents and prime ministers delivered their set-piece addresses to a sparsely populated hall, but contributions from two unlikely visitors in the nearby security council chamber stood out. Continue reading...

Across videos, standup and cartoons, humorists are skewering big tech’s attempt to make AI ubiquitous and inevitable “Best case scenario, we never have to do any admin again,” says an AI CEO. “Worst case, we all die.” A young man struggles to the top of a beautiful mountain, words echoing in his ears: “AI will most likely lead to the end of the world, but in the meantime, there will be great companies created.” At the top of the mountain, he triumphantly holds up a sign: “Build a datacenter here.” Continue reading...

Discussion | Link

Continue reading...

I suspect nearly all of my students use AI, and I worry about its corrosive effects. We must not entrust tech companies with our long-term health Prof Devi Sridhar is chair of global public health at the University of Edinburgh When I was a postgraduate student at the University of Oxford in the early 2000s, I interviewed for a major consulting firm. All my friends were being recruited, and I thought why not give it a shot? Aside from the usual maths-proficiency test (how quickly can you crunch numbers in front of a client and look smart), I was asked about what the future of streaming could look like. I gave a long-winded answer about the unintended consequences of changing how humans interact with each other when a new technology is introduced, and why this may not be in the best intere…

Discussion | Link

Discussion | Link

Discussion | Link

Discussion | Link

Discussion | Link

Xi Jinping’s visit heavy on symbolism but no sign closed-door meeting yielded breakthroughs on thorny issues such as AI, trade, Taiwan and Iran Donald Trump welcomed Xi Jinping to the White House on Thursday, with talks on AI, trade and a growing US-China strategic rivalry expected to feature on an elaborate state visit that critics say went big on style but lacked substance. Here are the key takeaways from Xi’s first visit to Washington in more than a decade. Continue reading...

Discussion | Link

Northern Gannet, Great Blue Heron, California Brown Pelican, in Monterey Bay National Marine Sanctuary, CA, US, CA New 200-800mm Canon EF lens got me my best photo of Morris yet. They really like hanging out under that sign in the harbor! Tags: photography, wildlife

Discussion | Link

Anthropic and OpenAI released new models this week, expanding the range of models enterprises can use for different workloads and raising new considerations for multi-model management.

Release: commit-rewriter 0.2 Support for branches other than the default branch. Use uvx commit-rewriter --branch other to run against another branch. #3 Tags: git

Discussion | Link

Discussion | Link

As Jensen Huang puts it, AI can help fight climate change - but only if it inflicts "an enormous amount of pain and suffering" first. The Nvidia CEO discussed the future of energy and AI's impact on our planet in the latest episode of The Ezra Klein Show. But his comments boil down to the same accelerationist spin we've seen from a variety of tech leaders as well as President Donald Trump. They promise that AI will be such a gift to humanity that it's worth the damage caused by data centers currently running on dirty energy. Huang's extreme level of privilege is glaring Huang's extreme level of privilege is glaring in the interview. He … Read the full story at The Verge.

Meta has a new plan to get people to make games for its Horizon social platform. The company today announced two new development tools that will let you create games with AI prompts: Horizon Create, a mobile app, and Horizon Studio, a browser app that offers more granular controls. The apps will be available in early access, and interested users can sign up for a waitlist. Horizon has struggled as a platform, but Meta has a plan to give these new Horizon games significantly more reach: It's going to let published games made with Horizon Create and Horizon Studio get recommended on Facebook and Instagram, where they also will be playable. … Read the full story at The Verge.

Datacentre hailed by government was supposed to launch next year but may be held back into mid-2030s A huge datacentre project hailed by the UK government will miss its launch date next year and could be delayed into the mid-2030s. The site in Loughton, Essex, was described as the country’s largest AI supercomputer when it was announced in 2025, but power supply problems mean it now faces a lengthy wait before coming online. Continue reading...

Discussion | Link

Continue reading...

I used Google’s Nest Doorbell (left), Aqara’s HomeKit-compatible G400 (center), and the Ring Pro 4K (right) to test the new AI features of Gemini for Home, Apple Intelligence for Home, and Amazon’s Ring. | Photo by Jennifer Pattison Tuohy / The Verge A few years back, I was at a beachside Easter egg hunt, watching my kids dash through sand dunes searching for sweet treats. My phone buzzed in my pocket; I ignored it. A moment later, it buzzed again. I pulled it out, glanced down, and saw a "motion detected" notification from my security camera. "Just my dog in the yard," I thought, dismissing it. As the colorful eggs piled up in the baskets, my phone kept buzzing. Reluctantly, I pulled it out again and saw several more "motion detected" alerts. I tried to load a video clip, but it just buf…

Chinese firms supplied nearly all dexterous hands for humanoid robots in early 2026, but Western players are starting to close in.

arXiv:2609.28816v1 Announce Type: new Abstract: Robotic bodies are inherently distributed in sensing and actuation, yet learning-based control still commonly relies on centralized information processing. This work studies the problem of information organization in communication-constrained embodied control: which computations should remain local, and which information is worth transmitting for whole-body coordination. We propose FlyCNS, an embodied information-organization framework inspired by the Drosophila brain--nerve-cord connectome. FlyCNS preserves local sensorimotor computation within each limb and enables selective long-range communication through separate ascending and descending routing pathways. From a real connectome, FlyCNS extracts the directional structural complexity of t…

arXiv:2609.28530v1 Announce Type: new Abstract: A general-purpose vision-language model can understand a task goal without knowing how a particular robot's motion and functional parts produce the intended effect. We introduce KnowBody, a harness that makes these action-relevant body relations explicit, queryable, and revisable while keeping the model weights frozen. Initialized from one off-task trajectory, a partial body model guides action selection and the interpretation of past interactions. New evidence refines the model, and knowledge dependent on revised body estimates is rechecked before reuse. Across 32 fixed-budget trials on four real-robot tasks, initialized KnowBody achieves 75% completion versus 25% for the native harness and requires fewer planner rounds on successful trials…

The AWS WhisperX Deep Learning Container packages Whisper, wav2vec2 forced alignment, and speaker diarization into a GPU-ready image. Learn how to deploy it to Amazon SageMaker AI real-time and asynchronous endpoints for word-level, speaker-labeled transcription, plus the production details that matter: the GPU AMI pin, scaling, and cost controls.