Quantum Zeitgeist · 12m ☆

Anthropic's Claude improves quantum optimization designs

Quantum Zeitgeist reports Anthropic's Claude improved quantum optimization designs.

The Decoder · 30m ☆

Mathematicians dispute OpenAI's Navier-Stokes proof methodology

Why it matters: Expert pushback on a headline AI math proof tests claims that models can do frontier research.

OpenAI submitted a 166-page Navier-Stokes solution, but expert groups say the actual approach differs from OpenAI's description. Because language models are black boxes, the reasoning can't be fully verified, The Decoder reports.

Wired · 31m ☆

Wired: AI lacks data to decode dynamic human genome

Wired argues DNA is far more dynamic than a static instruction set, and AI models still lack the training data needed to properly interpret it.

Livemint · 1h ☆

SBI warns customers about AI voice-cloning scams

India's State Bank warns customers about fraudsters using AI voice cloning to impersonate relatives and officials, and lists five ways to protect money and accounts.

The Hindu · 1h ☆

Anthropic bans cruel behaviour toward Claude

Anthropic says it bars persistently harmful or abusive use of Claude and has given the model the ability to end conversations in rare, extreme cases of abuse.

Pandaily · 1h ☆

China's Fengqing AI ensemble forecast system enters operation

Fengqing Ensemble V1.0 passed operational review at China's National Meteorological Centre, with an 11-day effective forecast range for global 500 hPa height in independent tests.

Pandaily · 1h ☆

Shanghai AI Lab releases Atria Dawn full version with 512K context

Why it matters: Advances large open agent models toward long-context agentic tasks.

The agent model built on a 744B MoE GLM-5.2 base moves past its September preview with higher scores on JobBench, WideSearch and PaperBench. Full weights remain marked 'coming soon'.

Cybersecurity Insiders · 2h ☆

Op-ed warns AI is collapsing cyberattack attribution

A Cybersecurity Insiders piece argues AI makes it easier for attackers to frame anyone, undermining cyber attribution capabilities.

Forkast News · 3h ☆

Forkast examines SEC Reg CA and AI agent liability gap

The piece argues SEC Regulation CA leaves unclear who is liable when autonomous AI agents transact, a gap for crypto and agent-driven finance.

Philosopher argues AI music generation insults musicians

A philosopher of music makes the moral case for human-made music, arguing that AI music-generation software disrespects the craft and labor of musicians.

RuntimeWire · 6h ☆

Anthropic gives METR staff internal access to assess frontier risks

Anthropic granted nonprofit METR six staffers internal access to assess frontier AI risks, letting METR publish assessments with limited redactions; Anthropic is also pitching lawmakers on a draft AI safety bill.

The Hindu · 6h ☆

Kevin Rudd warns US, China of AI security crisis

Former Australian PM Kevin Rudd said autonomous AI systems able to infiltrate or attack digital infrastructure pose major risks to national security, financial systems and essential services, and called for US-China cooperation.

Bioengineer.org · 7h ☆

Humanoid robot speaks Quechua via fine-tuned neural speech synthesis

Researchers fine-tuned a neural speech synthesis model so a humanoid robot can speak Quechua, an Indigenous Andean language, demonstrating low-resource language TTS on a robot.

Mint · 8h ☆

Google's SynthID Detector checks images, video and audio for AI marks

Why it matters: Gives a mainstream tool for spotting AI-generated media, though gaps remain.

Google's SynthID Detector scans images, videos and audio for invisible AI watermarks, giving professionals a first verification step. A missing watermark does not prove content is authentic.

TronWeekly · 8h ☆

Sui integrates Alibaba Cloud into agent payments

The Sui blockchain integrated Alibaba Cloud into its agent payments stack, aiming to let AI agents transact on-chain with cloud infrastructure support. SUI price watched at $1.40.

OrcaRouter · 8h ☆

Nvidia Nemotron Labs releases Dory 1.0 4B on Hugging Face

Dory 1.0 4B is a small 4-billion-parameter model quietly released on Hugging Face by NVIDIA Nemotron Labs; a comparison pits it against Qwen 3.8's three-tier lineup.

CyberScoop · 8h ☆

US agencies recommend cyber controls, testing for frontier AI

Why it matters: Sets baseline federal cyber expectations for testing frontier models.

Federal agencies issued voluntary cybersecurity guidance for securing sandbox environments during frontier AI model tests, the government's most detailed such recommendations, including independent testing.

TechBullion · 9h ☆

TechBullion examines voice AI and real-time interpretation at work

The piece describes how voice AI and real-time translation tools are being used across language barriers in global enterprises, framing interpretation as a borderless business capability.

forkast.news · 9h · 3 sources ☆

Six vendors ship agent identity products as IETF standard lags

Why it matters: Shows agent identity tooling outpacing standards and regulation.

Cequence and five other vendors are shipping commercial agent identity layers while the IETF has yet to finish the standard, leaving legal and interoperability rules unclear.

IndieWire · 10h ☆

Artlist tests AI-made sci-fi series 'The Arc' for $38,000

9to5Mac · 11h · 5 sources # ☆

Apple adds App Store rules targeting AI nudify and deceptive apps

Why it matters: Platform rules from Apple shape how AI image and deepfake apps reach iPhone users.

Apple updated its App Review Guidelines, adding rules against AI nudification and 'bait-and-switch' apps, following a deepfake scandal involving Grok.

디지털투데이 · 11h ☆

AI agent 'wallet wars' shift from virtual cards to agent tokens

Why it matters: Agent identity and payment credentials are the sticking point for agentic commerce.

Analysis says agentic payment infrastructure is moving away from virtual card numbers toward dedicated agent tokens for AI agents that transact.

Minonline · 11h · 2 sources ☆

xAI's Grok Voice 3.0 targets phone agents at $0.08/min

Why it matters: Puts xAI into the fast-growing market for cheap real-time voice agents.

Grok Voice 3.0 answers in about 0.81 seconds and topped Artificial Analysis' conversational voice rankings with an 86.9 score and 94.8% task success rate, priced at $0.08 per minute for AI phone agents.

Trend Hunter · 12h ☆

Square launches voice-enabled self-service ordering kiosks

Square introduces conversational AI ordering kiosks that let customers place self-service orders by voice.

Reddit · 12h ☆

Claude Haiku 4.5 to leave model selector on October 29

Users report that Claude Haiku 4.5 will be removed from the Claude model selector on October 29, prompting discussion among users who rely on the faster model.

WSJ · 12h ☆

WSJ: AI companies play cat-and-mouse with spies

Why it matters: Model IP theft is now a top security concern for frontier AI labs.

The Wall Street Journal reports AI labs are countering espionage attempts targeting model secrets, escalating security measures as rivals seek proprietary methods.

American Banker · 13h ☆

Colony Bank uses Lama gen AI to speed SBA loan approvals

Georgia's Colony Bank is using software from Lama to originate small-business loans, cutting days off SBA loan approval times.

The New York Times · 13h · 2 sources ☆

Terence Tao joins debate on AI and the future of math

The New York Times profiles mathematician Terence Tao's role in the debate over AI's impact on mathematics, including his criticism of dumping AI-generated math solutions on others.

OrcaRouter · 13h ☆

OrcaRouter compares Xing 4.0 29B-A4B with Qwen 3.8 27B

OrcaRouter published comparisons of vision and tool-calling between Xing 4.0 29B-A4B and Qwen 3.8 27B, and of Claude Mythos 5 and Kimi K3 on two containment failures.

RuntimeWire · 13h · 2 sources ☆

OpenAI adds parallel agents to GPT-6 Pro in ChatGPT

Why it matters: Parallel delegated agents in ChatGPT push the assistant toward multi-task autonomy.

OpenAI rolled out a Chat tab for Pro users to track delegated tasks handled by parallel agents across plugins and tools, though it gave no latency figures or examples.

RuntimeWire · 13h ☆

Pika brings its image, video and sound toolkit into ChatGPT

Pika's creative tools are now integrated into ChatGPT, adding a side panel, timeline editing and access to users' Pika libraries for image, video and sound generation.

PYMNTS AI · 13h · 3 sources # ☆

Taco Bell deploys voice AI at over 890 drive-thrus

Taco Bell says voice AI now takes drive-thru orders at more than 890 locations, handling order-taking that staff previously did over headsets. The chain says the system helps with retention.

Raleigh News & Observer · 13h ☆

Explainer covers what AI shopping agents do

The piece explains how AI shopping agents act on a user's behalf when buying products, covering their capabilities, limits and risks.

Dark Reading · 13h ☆

Fake Claude and ChatGPT campaigns surge to target developers

Why it matters: Brand-impersonation attacks on AI tools show how AI adoption expands enterprise phishing surface.

Dark Reading reports a rise in campaigns impersonating popular AI brands like Claude and ChatGPT, using ClickFix-like tactics to target unsuspecting enterprise employees and developers.

Reddit · 13h ☆

ORL speeds diffusion LLMs to 2,788 tokens/s on one RTX 5090

Why it matters: Shows diffusion language models can be served at interactive speeds on consumer GPUs.

ORL (On-Policy Residual Learning) is an open-source method that makes diffusion language models generate much faster without losing quality; with the custom dlmrt inference engine it hit 2,788 tokens/s on a single RTX 5090.

La Voce di New York · 14h · 4 sources ☆

NY Speaker Menin confronts xAI after Grok made sexualized image of her

Why it matters: Highlights the ease of nonconsensual AI image generation and deepfake harms.

New York City Council Speaker Adrienne Menin says xAI's Grok generated a sexualized image of her in under 30 seconds; she confronted company representatives over the nonconsensual imagery.

The Hindu · 14h · 2 sources ☆

OpenAI to watermark ChatGPT text in EU under AI Act

Why it matters: Watermarking ChatGPT text sets a compliance precedent for AI content labeling in Europe.

OpenAI, a signatory to the EU's AI-content code alongside Google, Anthropic, Meta, Microsoft and Mistral, will watermark ChatGPT text in the EU; tests show the mark can weaken sharply after editing.

Palo Alto Networks · 14h · 2 sources ☆

Palo Alto Networks launches 5th-gen firewalls built for AI scale and quantum

Palo Alto Networks introduced its fifth-generation hardware firewalls designed for AI-scale workloads and quantum resilience, alongside its expanding AI security portfolio.

Yahoo Finance · 14h ☆

Bug in OpenAI's Office highlights AI espionage risk, Yahoo Finance says

A bug in OpenAI's Office is examined as a case study in how AI systems could be exploited for espionage, per a Yahoo Finance column.

MediaPost · 14h ☆

Breztri runs AI-created llama ads breaking pharma tropes

Breztri's new campaign uses AI-generated llama imagery to upend pharma advertising conventions, per MediaPost.

www.marketscreener.com · 14h · 2 sources ☆

Kuaishou launches Kling 4.0 video generation model

Why it matters: Kling is a leading Chinese video model, and version 4.0 raises the bar for AI video generation.

Kuaishou launched Kling 4.0, the next generation of its Kling AI video generation model, amid investor questions over whether the company's shares are undervalued.

Dark Reading · 14h ☆

AI takes control of robotic arm in 16 minutes, Dark Reading finds

Why it matters: Demonstrates physical risk when AI-enabled industrial robots are left under-protected.

Dark Reading reports an attacker gained control of an industrial robotic arm within 16 minutes, highlighting weak security in long-deployed OT devices at utilities and factories.

Reddit · 15h ☆

Plenio 1.0 launches free AI song studio for ComfyUI

Plenio Music Production System 1.0 is a free AI song studio for ComfyUI that takes a track from first idea to finished song, now with a player that makes videos, visuals and shareable mixtapes.

Condia · 15h ☆

Condia: doctor typing speed highlights Africa voice AI gap

Condia argues Africa's voice AI gap is visible in how slowly doctors type clinical notes, making speech interfaces a major opportunity for local health systems.

Incode launches TrustGraph trust network for agentic era

Why it matters: Verifying AI agents as trusted actors is a growing need as they transact on users' behalf.

Incode released TrustGraph, described as the first trust network built for the agentic era, aimed at verifying identities of people and AI agents in digital transactions.

Blueport adds AI agent readiness to Essentials platform

Blueport brings server-side rendering and AI agent readiness to its Essentials e-commerce platform, aiming to make retailer sites readable and actionable for shopping agents.

HackerNoon · 15h ☆

MemeToro open-sources its AI trading agent

MemeToro has open-sourced its AI agent for meme coin trading, letting developers inspect and run the code behind its automated crypto trading approach.

MIT Technology Review · 15h ☆

MIT panel discusses AI-designed virus research

MIT Technology Review hosted a roundtable with Samuel King, the Stanford PhD student who used a generative AI model to propose genetic blueprints for microscopic viruses. The work is preliminary, not yet AI-generated life.

Reddit · 15h ☆

Developers release Tyger open-weights streaming ASR with turn-control

Tyger is a small voice-agent model built on Nvidia's Nemotron 3.5 ASR Streaming 0.6B plus a 14M causal GRU that picks answer, stop or wait every 80 ms with no lookahead.

CSO Online · 16h ☆

Hallucinated domain in OpenAI skill highlights AI supply chain risk

CSO Online reports that a hallucinated domain inside an official OpenAI skill underscores supply chain risks for AI agent tooling.

Article warns of AI-enabled deepfake and voice-cloning scams

A fraud-awareness piece explains how scammers use AI deepfakes and voice cloning to impersonate people and trick victims into sending money.

quasa.io · 16h ☆

Listen2It launches AI audio suite with voiceover and cloning

Listen2It is an AI audio suite for generating voiceovers, multi-track studio editing and voice cloning from text, aimed at podcasters and creators.

The Times of India · 16h ☆

Two held for extorting Ramleela committee chair with AI video

Two people were arrested in India for allegedly extorting the head of a Ramleela organizing committee using an AI-generated video, showing deepfake-enabled blackmail.

American Banker · 16h · 3 sources ☆

M&T Bank monitors impact of AI agents on deposits

M&T Bank says it is tracking how AI agents used by customers affect deposit flows, as US banks weigh how agent-driven account access could change balances and money management.

Hacker News · 16h · 2 sources ☆

Mercor estimates 90% of US work hours lack AI benchmark coverage

Why it matters: Highlights a large evaluation gap between AI benchmarks and real work.

Mercor research finds roughly 90% of US working hours are spent on tasks current AI benchmarks do not cover well, using a labor index benchmark map.

The Next Web · 16h ☆

MI5 opens AI threat centre as OpenAI warns UK MPs

Why it matters: Signals intelligence agencies treating AI-enabled threats as a standing priority.

UK intelligence service MI5 opened an AI threat centre, while OpenAI told MPs that more AI-related incidents may come. The moves reflect growing government focus on AI-enabled security risks.

IndieWire · 16h ☆

Alex Gibney used AI to simulate an Elon Musk interview

Why it matters: AI-generated interviews in documentary film rekindle debates over synthetic media and journalistic ethics.

Documentary filmmaker Alex Gibney describes using AI to recreate and simulate an Elon Musk interview he could not obtain, illustrating AI's use in documentary production.

Winston-Salem Journal · 16h ☆

WSSU faculty win federal grant for AI cybersecurity work

Faculty at Winston-Salem State University received a federal grant to develop protections against AI-driven cybersecurity threats, funding research and training at the university.

SiliconANGLE · 16h · 2 sources ☆

Raymond James uses Neo4j knowledge graphs to ground AI agents

Raymond James Financial uses Neo4j knowledge graphs to evaluate and ground AI agent results, as knowledge graphs become a context layer for agentic AI in financial services.

FinancialContent · 16h ☆

Report: gift cards emerge as a gateway to agentic commerce

A report argues gift cards could become an early on-ramp for AI agents that shop and pay, letting agents transact without full card credentials.

Bioengineer.org · 2h ☆

Artificial skin with brain-like wiring gives robots human touch

Why it matters: Better tactile sensing is a key missing piece for dexterous humanoid robots.

Researchers describe an artificial skin whose wiring mimics brain-like signaling, giving humanoid robots a more human sense of touch for manipulation and interaction.

超能网 · 3h ☆

Nvidia appears to drop RTX 5090 from official sales platform

Nvidia appears to have removed the RTX 5090 from its official sales platform, with retailers also gradually clearing out inventory of the flagship AI-capable GPU.

RuntimeWire · 4h ☆

Dynamic Creatures launches Hugo guest-interaction robot

Marc Theermann's startup Dynamic Creatures introduced Hugo, a mobile robot for guest interactions in theme parks and hospitality, using a Vision-Language-Expression model to connect perception to movement.

igor´sLAB · 5h ☆

Nvidia releases cuPotential GPU library for atomic simulations

Nvidia's cuPotential is a GPU library that accelerates atomic-scale simulations used in semiconductor manufacturing.

Bioengineer.org · 6h ☆

ISEED robot learning system develops skills like a baby

Researchers describe ISEED, a robot learning approach that lets machines acquire skills through development rather than from a static dataset, similar to how infants learn.

Pandaily · 7h ☆

UBTECH details Thinker-WM embodied world model

Why it matters: Shows depth conditioning improves embodied world models for robot manipulation.

UBTECH's Thinker-WM unifies video, depth and robot actions; it tops three embodied world model benchmarks, and adding depth lifted single-task RoboTwin success by 15 points in ablations.

Pandaily · 7h ☆

Rokid and Tencent Mobility add voice ride-hailing to AI glasses

Why it matters: Puts agentic booking and payment into a wearable, a concrete smart-glasses use case.

Users say a destination and Rokid glasses handle pickup points, fare comparison, booking and password-free payment via an agent-to-agent link with Tencent Mobility, showing car details while waiting.

Pandaily · 7h ☆

Unitree releases free UniUni swarm control software for humanoids

Unitree's UniUni is a Windows and macOS app to draw paths, arrange timelines, sync motion to music and monitor a humanoid fleet with one-tap emergency stop; free for its users, starting with H2 and G1.

Growth Dragons · 7h ☆

Huawei China AI chip share rises to 28% against Nvidia's 34%

Huawei's share of China's AI chip market rose to 28% while Nvidia holds 34%; YMTC is now the world's No. 3 NAND maker, and Alibaba launched AI glasses, Growth Dragons reports.

RuntimeWire · 8h ☆

Neuralwatt says cache misses can raise inference energy 140%

Neuralwatt CEO Chad Gibson analyzed Neuralwatt Cloud traffic and found prompt edits cause wasted GPU work, with cache misses using up to 140% more inference energy, worst in long agent sessions.

freeyork · 8h · 3 sources # ☆

Amazon unveils always-listening Bee wearable and companion AI device

Why it matters: Amazon's always-listening wearable pushes ambient AI memory capture into mainstream consumer hardware.

Amazon introduced new AI devices, including the Bee wearable, designed to continuously capture and recall conversations without a screen.

DigiTimes Asia · 9h ☆

Tesla Roadster reveal fuses SpaceX rocket tech into physical-AI halo

Tesla unveiled its next-generation Roadster at SpaceX's McGregor, Texas test site on October 15, combining SpaceX and Tesla engineering as a showcase product for Musk's physical-AI ambitions.

Opportunity charging gains ground as robot fleets scale

Sparks' Ben Waters explains why frequent, distributed opportunity charging is becoming essential to keep growing autonomous robot fleets running without downtime.

PBS · 10h ☆

PBS examines hopes and risks of home humanoid robots

The segment weighs the promise of humanoid robots doing household tasks against safety, cost and reliability concerns before they enter ordinary homes.

Yonhap · 11h · 4 sources # ☆

North Korea tests AI-guided precision torpedo, KCNA says

North Korea said it conducted an underwater test of a new precision-guided torpedo equipped with AI capabilities, according to state media KCNA.

TweakTown · 12h ☆

HelixSR mod unofficially adds DLSS 3.7 to Nvidia GTX 10 GPUs

A community mod called HelixSR brings unofficial DLSS 3.7 support to Nvidia's older GTX 10-series graphics cards, which never officially received it.

Bioengineer.org · 13h ☆

Virtual forces help humanoid robots master hard moves

Researchers used temporary virtual forces to help humanoid robots learn and execute their most difficult maneuvers.

i-PRO launches edge-AI remote head camera

i-PRO introduced a new remote head camera that extends edge AI, letting analytics run on the device rather than in the cloud. It targets surveillance and industrial video deployments.

Aviation A2Z · 15h ☆

British Airways bans Meta AI glasses on flights

British Airways bans Meta AI glasses and threatens to remove passengers who film crew without consent, citing privacy and recording concerns.

MassRobotics opens Healthcare Robotics Startup Catalyst applications

MassRobotics is accepting applications for its Healthcare Robotics Startup Catalyst Program. The cohort runs February through June 2027 and ends with showcases at the Robotics Summit and Expo in Boston.

Embedded Computing Design · 16h ☆

electronica 2026 IIoT forum to cover edge AI and robotics

The IIoT Forum at electronica 2026 will examine edge AI, security and industrial robotics, highlighting where AI runs on factory hardware.

ZDNET · 16h ☆

iPhone Duo preorder shipments slip to late January

Apple's iPhone Duo preorders now show a ship date of late January, a new wait-time record for Apple and its customers, per ZDNET.

Facebook · 17h ☆

NYT profiles workers wearing cameras to train robots

More American workers are strapping cameras to their bodies to collect video data used to train robots, the New York Times reports, examining what this means for the future of work.

TechCrunch · 17h · 4 sources # ☆

Ex-Nvidia robotics executive says physical AI is stuck in 'Lego era'

Amulya Vishwanath, who spent seven years leading product marketing for Nvidia's robotics platforms and is now a VC, says robotics remains in a 'Lego era' with actuators, hands, software and safety standards still being rebuilt, and has not reached its ChatGPT moment.

EFE Comunica · 17h ☆

Humanoid robots improve but still struggle to stop moving

A report says humanoid robots have advanced in capability yet still lack reliable ways to halt motion, a basic safety gap for physical AI deployment.

Gadget Review · 17h ☆

AI shelf-scanning robots roll out at LA Ralphs stores

AI-powered shelf-scanning robots are now operating in Ralphs grocery stores in Los Angeles, checking inventory and pricing on shelves.

GlobeNewswire · 17h · 2 sources ☆

AiMOGA to launch Mobo home companion robot with global preorders

AiMOGA will launch Mobo, a home companion robot, and open global preorders at the 2026 Chery International User Summit.

The Register · 18h ☆

NTT: permits, not parts, could cap AI datacenter boom

In a sponsored post, NTT Global Data Centers says permitting bottlenecks, not equipment shortages, are the likely constraint on building datacenters fast enough for AI demand.

RuntimeWire · 19h ☆

XREAL 1 Pro+ adds 1,200-nit brightness to $599 glasses

XREAL's 1 Pro+ display glasses cost $599 and keep the One Pro's chip and 57-degree field of view, adding 1,200-nit brightness, AI HDR, faster 2D-to-3D conversion and a redesigned nose pad.

HotHardware · 19h ☆

Nvidia GeForce NOW adds 4X frame gen and DLSS 4.5 controls

Nvidia's GeForce NOW cloud gaming service now supports 4X frame generation with full DLSS 4.5 controls, letting subscribers tune upscaling and frame-gen settings per game via streaming.

Korben · 19h ☆

Remote-controlled Unitree R1 robot assaults teenager in Cologne

Why it matters: An assault by a commercially sold humanoid highlights physical-safety and remote-control risks as cheap robots spread.

Korben reports a remote-controlled Unitree R1 humanoid robot assaulted a teenager in Cologne, raising fresh questions about safety and control of commercially available robots.

Interesting Engineering · 19h ☆

Two humanoid robots play table tennis doubles

Why it matters: Coordinated multi-robot play with fast reaction is a hard physical-AI benchmark.

Two humanoid robots teamed up for a table tennis doubles match, rallying for more than ten shots in a demonstration of coordinated manipulation and reaction.

semivision · 20h ☆

Qualcomm CEO predicts AI phone era shifting from apps to agents

Qualcomm's chief executive forecasts the next technology revolution will move smartphones from app-centric use to AI agents that act on the user's behalf.

eeNews Europe · 21h ☆

Siemens and Nvidia speed up chip lithography 12x

Why it matters: Cuts lithography simulation time, a key bottleneck in advanced chipmaking.

Siemens and Nvidia demonstrated a 12x speedup in AI chip lithography by applying GPU-accelerated computing and AI to lithography simulation and mask optimization.

The Verge · 21h · 13 sources # ☆

Nvidia RTX Spark laptops go on sale starting at $2,599

Why it matters: Nvidia's first laptop superchip pushes local AI inference into mainstream consumer PCs.

Nvidia's RTX Spark N1X Arm-based superchip for laptops and small desktops is now shipping in devices like Microsoft's Surface Laptop Ultra, starting at $2,599. It runs AI models locally and rivals Apple and Intel silicon.

NeoTeo · 21h ☆

Sipeed details MaixStation's Qwen performance claims

Sipeed detailed the Qwen inference performance it claims for MaixStation, its edge AI hardware, in a NeoTeo report on the device's local model benchmarks.

Business Standard · 21h ☆

Experts weigh AI wearables' prospects after early setbacks

Business Standard asks whether AI wearables can win over buyers after early product failures, gathering expert views on privacy, usefulness and the path to mainstream adoption.

Robohub · 21h ☆

Robot Talk Episode 166 features Rika Antonova on robot co-design

Robot Talk Episode 166 interviews Rika Antonova of the University of Cambridge on using AI to co-design complementary robot hardware and algorithms and automated algorithm discovery.

The Washington Post · 22h ☆

Japan turns to AI robots and drones against bear attacks

As bear attacks rise, Japan is deploying AI, four-legged robots and pepper-spraying drones to detect and deter bears, the Washington Post reports.

MarketWatch · 22h · 2 sources ☆

Reporter spends afternoon with two $55,000 humanoid housekeepers

A MarketWatch reporter tested two humanoid housekeeping robots priced above $55,000, including whether they can load a dishwasher, finding current home-robot capability and cost limits.

MarketWatch · 22h ☆

Japanese construction firms turn to humanoid robots for labor gaps

Yomiuri reports Japanese construction firms are looking to physical-AI-powered humanoid robots to fill labor shortages.

Muks Robotics wins ISRO project to build SPACEO humanoid

Pune-based Muks Robotics has secured an ISRO project to develop SPACEO Prime, a humanoid robot intended to work in space. No timeline or value disclosed.

Vietnam.vn · 23h ☆

Dancing humanoid robots and robot dogs draw crowds in China

A Vietnam.vn report highlights Chinese humanoid robots and quadruped robot dogs performing dance routines, drawing public attention as the country pushes physical AI and consumer robotics displays.

Tom's Hardware · 23h ☆

Ukrainian drones strike Russian data centers, disrupting services

Why it matters: Drone strikes on data centers show AI infrastructure is now a wartime target.

Ukrainian long-range drones hit Russian data centers, knocking out or severely disrupting Russian digital services and apps. Some reports suggest online bot activity also fell, though less than first claimed.

China Daily · 23h · 2 sources ☆

Astribot brings T1 humanoid to North America

Astribot is bringing its T1 humanoid robot to North America, showcasing a physical AI stack built around its 'Design for AI' approach.

The Daily Advertiser | Lafayette, Louisiana · 1d · 4 sources ☆

Blind veterans use Meta AI glasses at races and Milan museum

A blind Louisiana veteran used Meta AI glasses to complete an Army race, other blind veterans received free pairs, and the glasses now describe eight Brera masterpieces for blind visitors in Milan.

Robotics & Automation News · 1d ☆

Tesla patent reveals touch-sensitive skin for robot hands

Why it matters: Tactile skin patents show humanoid makers racing to close the dexterity gap.

A Tesla patent describes touch-sensitive skin for robot hands, per Robotics & Automation News. The filing covers tactile sensing likely for Optimus humanoid hands.

The New York Times · 1d ☆

NYT: fashion brands cozy up to Meta glasses and AI tech

The New York Times reports fashion companies are increasingly collaborating with tech firms on Meta smart glasses, AI advertising and robot models as the industries converge.

IDTechEx overviews sensor technologies in humanoid robots

IDTechEx published a video overview of sensor technologies used in humanoid robots, covering the sensing hardware that enables balance, touch and perception.

AiThority · 1d ☆

RLC Pro AI runs on Nvidia DGX Spark with Enterprise Linux

RLC Pro AI says it delivers top AI performance on Nvidia's DGX Spark desktop AI computer while keeping Enterprise Linux stability.

Reddit · 1d ☆

Maker builds $200 hexapod robot from old Android phone

A hobbyist built a hexapod robot powered by a repurposed Android phone, costing under $200, and describes it as a good first robotics project for learning.

Reddit · 1d ☆

Artificial Analysis launches AA-AgentPerf benchmark for inference hardware

Why it matters: Agent workloads stress inference hardware differently, making hardware benchmarks for agentic use increasingly important.

AA-AgentPerf benchmarks inference hardware by replaying real coding-agent trajectories with long contexts, tool-call pauses and KV-cache reuse, plotting server throughput per megawatt against per-request output speed. Vera Rubin leads energy efficiency.

СПЖ · 1d ☆

AI-powered robot monk built in Japan to 'calm souls'

A Japanese project created an AI-driven robot monk, intended to offer spiritual comfort and conversation to visitors, according to a report.

NVIDIA · 1d · 6 sources # ☆

Nvidia open-sources ASPIRE skill-discovery system for robots

Why it matters: An open robotics skill-learning stack could speed physical-AI generalization.

ASPIRE is Nvidia's open-source agentic skill-discovery system that lets robots build reusable skills from experience; Nvidia says it taught a robot to solve a Rubik's Cube.

Pandaily · 1d ☆

WanNa Robotics debuts NOVA21 direct-drive dexterous hand

Why it matters: Higher-DoF direct-drive hands advance dextrous manipulation for humanoids.

NOVA21 is a robotic dexterous hand with 21 active rotary joints and 16 force sensors, shown at the Hong Kong Autumn Electronics Fair with industrial fieldbus options.

Pandaily · 1d ☆

AgiBot places YuanSheng robots at 20+ Chinese universities

Why it matters: Campus deployments give AgiBot large-scale real-world robot data and talent pipelines.

AgiBot's YuanSheng program will put Genie and Lingxi humanoids, Expedition and quadruped robots plus AIMA software seats at more than 20 Chinese universities, over 1,000 robots in total.

DigiTimes Asia · 1d ☆

Chunghwa Telecom builds Taiwan network for AI agent era

Chunghwa Telecom president Rong-shy Lin said the shift from conversational AI to autonomous agents is reshaping enterprise adoption, making data governance, access controls and accountability critical at scale, as the telco builds Taiwan's high-speed computing network.

Robotics & Automation News · 1d · 2 sources ☆

Pudu expands from service robots into quadrupeds

Pudu Robotics is expanding beyond service robots into semi-humanoid and quadruped robot lines, extending its commercial robotics lineup into new form factors.

AiThority · 1d · 3 sources ☆

Skild AI trains robots to learn physical skills via self-play

Why it matters: Self-play training for physical skills pushes humanoids toward autonomous multi-step manipulation.

Skild AI's robot brain lets humanoids learn physical skills through self-play and handle multi-step tasks autonomously. Cyera's AI security platform also launched in the Claude Marketplace.

TechNode · 1d ☆

Hesai's Kosmo turns real places into 3D assets for robots

Why it matters: Physical AI's biggest bottleneck is real-world data, and 3D asset capture aims to fill it.

Kosmo, from lidar maker Hesai, captures real-world locations and converts them into 3D assets robots can learn from, addressing the shortage of physical-world training data for embodied AI.