This reported breakthrough apparently used a dataset that’s open... for researchers. It’s the work of Eddy Xu, a teenager who was one of the first to get in on the video training data gold rush. He dropped out of Columbia last year to launch Build AI, which has raised around $22 million so far. Build AI’s Egocentric-1M dataset reportedly includes 1 million hours of data recorded using the startup’s self-developed devices across factories in Southeast Asia. A lot of it is from India. Xu has said he’s moved his team to Bengaluru, dedicating $10 million to get data from Indian factories. India has become one of the prime locations for collecting this kind of data. While enrolled at Columbia Engineering, Xu went viral in January 2025 after showing Meta Ray-Ban smart glasseshe modified to cheat at chess. The student, then 17, connected the device’s camera to a chess engine that calculated the best move and relayed it in real-time. The tech reached a wider audience thanks to popular streamer and chess master Alex Botez publicly tested them. Before college, the Long Island-raised Xu won DECA’s global business championship and sold an edtech startup that reached more than 178,000 users in 90 days. He also launched a startup called Omega Robotics in middle school, raising about $120,000 to run an independent, coach-free competitive robotics team out of a basement. Xu and co-founder Jonathan Jia, who serves as CTO, moved to San Francisco to build the first recording devices with a small team. They quickly moved operations to Shenzhen to quickly iterate and scale production. Build previously offered smaller datasets with 10,000 and 100,000 on Hugging Face but the 1M dataset requires emailing Xu directly. I’m sure he’s flooded with requests now.show more

‼️ California metallurgist David Xu, who pleaded no contest... to poisoning a colleague Rong Yuan, is seeking to have his felony conviction dismissed, years after hidden-camera footage exposed him contaminating her water bottle with toxic heavy metal cadmium ‼️ David Xu, a metallurgist at Berkeley Engineering and Research, spiked his colleague Rong Yuan’s food and water with cadmium — a toxic heavy metal — for nearly 18 months between 2017 and 2019. As Yuan suffered debilitating symptoms, hospital visits and fear of a terminal illness, Xu maintained a deceitful facade, even suggesting her illness was caused by “mold” in the office. His poisoning campaign only unraveled in early 2019 when Yuan set up a hidden camera at her desk, capturing Xu in the act of tampering with her water bottle. The video evidence led to Xu’s arrest in Mar 2019 on charges of attempted murder and poisoning, which also affected Yuan’s parents, who had inadvertently consumed the poisoned water. Xu faced up to life in state prison. After his arrest, Xu was released on $2 million bail. He then received repeated modifications to his home confinement, with his travel restrictions eventually fully lifted, A judge set bail at $2 million, and Xu was released on GPS monitoring. In April, Attorney General Xavier Becerra filed a motion on behalf of the state engineering board, calling Xu “a highly trained and unusually sophisticated criminal defendant” who “used a number of chemicals in his work.” A judge agreed with the AG’s order and barred Xu from practicing under his state engineering licenses during the case. In the ensuing months, the court granted Xu more and more leniency. By July, he was allowed 6 hours out of the house each week to take his daughter to ballet and gymnastics classes. In Oct, he received 6 more hours out of the house on Sundays. In June 2021, he was granted “an additional day with kids.” By Sep, he was allowed an additional 2 hours every day of the week. In Nov, his request to end house arrest was granted. By Apr 2022, he was allowed to travel within the U.S. In Sep 2023, Xu pleaded no contest to a single count of poisoning under a deal offered by prosecutors under Alameda County district attorney Pamela Price, with one year of felony probation and one day of credit for time served. A no-contest plea carries the same weight as a conviction but is not an admission of the underlying facts. Yuan had sought nearly $500,000 in restitution. A judge ordered her paid $7,058, citing a report Xu’s attorney had submitted by a toxicologist who concluded that cadmium poisoning would not cause long-term health effects. Xu finished his two years of probation in March. On Oct 12, he will ask an Alameda County judge to dismiss his felony conviction — a process commonly referred to as “expungement.” Under state law, a person who completes every condition of probation is entitled to expungement, and California appeals courts have held that judges cannot deny those petitions unless the crime is on a short list of exclusions — poisoning not being one of them. A dismissal would not erase the case; Xu’s record would still show an arrest and conviction to law enforcement, and the conviction could be used against him if he is charged with a new crime. He would also have to disclose it on applications for state licenses.show more

Pi Ventures Joins Hack VC To Back AI Robotics... Startup On Base Axis Robotics (Axis Robotics) has raised a $12 million seed round led by Hack VC, with participation from Pi Network (Pi Network) Ventures, Nomad Capital, 10K Ventures, and other angel investors. The company is building a data engine for Physical AI that combines simulation, real world data capture, and human feedback to generate scalable robotics datasets. Axis said the funding will accelerate development of its global human in the loop data engine. Base (Base APAC) congratulated the team, calling Axis one of the leading scalable Physical AI and robotics platforms building on the network.show more

be Alex Mashrabov > start programming at 10 >... get a Meta internship offer at 18... turn it down > build an AI startup called AI Factory instead > Snap buys it for a reported $166 million in 2019 > leave and co-found Higgsfield to launch the AI video platform in 2025 > get 32 million users and nearly 1 million paying subscribers > on track for $1 BILLION+ in annualized revenue > built one of the fastest-scaling gen AI companies in the US and he says they're barely scratching the surfaceshow more

Humanoid robots have a truly insatiable appetite for data.... Midcentury, another New York-based physical AI company, claims to have surpassed the milestone of 2 million hours of human first-person behavioral data: > It covers over 50 environments and 20,000+ tasks, with hands clearly visible in more than 90% of the footage. > It offers more than just raw video; the dataset includes 3D hand tracking, depth maps, and task annotations, and even bundles in 50,000 hours of gameplay data and 69,000 hours of multilingual dialogue. > Beyond selling data, they have developed a cloud-based simulation platform called "Matrix," enabling humanoid robots to train directly in the cloud using this data. --- Companies like Dyna Robotics have demonstrated that if you feed AI enough egocentric/human-centric data, it can independently grasp the objective laws of the physical world, yielding astonishing model performance. It is fair to say that human behavioral data is becoming the "new oil" of the physical AI era. This achievement has helped Midcentury secure $15 million in seed funding.show more

🚨 CRYPTO: OKX CEO STAR XU AND BINANCE FOUNDER... CZ CLASH PUBLICLY OVER BOOK CLAIMS AND PERSONAL HISTORY OKX CEO Star Xu (Star_OKX) publicly challenged Binance founder CZ's iconic origin story of selling his home to buy Bitcoin, claiming the narrative omits critical family involvement and that the property belonged to CZ's in-laws. Xu also accused CZ of falsehoods in his recently published book and challenged him to produce a signed divorce agreement. CZ 🔶 BNB fired back directly, confirming he is "officially divorced" and offering to bet $1 billion with lawyers verifying the documents. He gave Xu a 24-hour window to accept the bet, saying failure to respond "clearly shows who has been misrepresenting to the public." CZ declined to post legal documents publicly out of respect for his ex-wife's privacy. The exchange marks a rare and public escalation of the long-running rivalry between the founders of the two largest crypto exchanges, with posts from both sides pulling 300,000+ views within hours.show more

🚀 Introducing EgoExo Forge - built on top of... Rerun, Gradio, and Hugging Face hub (I’ll be in San Francisco July 21–29 — if you’re into robotics, egocentric AI, large-scale data collection, or just want to chat, DM me!) In my opinion, large-scale, diverse, and high-quality data is still the largest bottleneck for generalized robotics deployment. I believe that some version of imitation learning from human examples will be the most scalable + clean way to train humanoid robots 🤖 (similar to what Tesla did for Full Self Driving). Teleop is too expensive to collect a large enough dataset in a reasonable manner, so passive collection via egocentric (and in certain cases, exocentric) views feels like the right bet. Over the past few months, I've been trying to build out the scaffolding for this and using Rerun as my underlying infrastructure. Data being collected needs to be easily inspectable + time series and rerun provides the right tooling for this. My goal is to first build out a ground truth representative dataset from already existing open source data, generate some reasonable baselines, and then go out and collect my own data that adheres to the defined schema. 🔍 Starting with open-source datasets 1. EgoDex from Apple 2. HOCap from Nvidia and the University of Texas at Dallas 3. Assembly101 from Meta All these different datasets have different sensor configurations + annotations, so my goal with egoexo-forge is to have one consistent labeling scheme + data layout. I built a data pipeline that aligns all of the different datasets in one general schema assuming the COCO133 keypoint layout that allows for exo+ego, ego only, or exo only Since the scaffolding is already there, it becomes MUCH easier to add other datasets. So the next ones that I'll be including are HD-EPIC kitchens dataset, HOT3D, and finally my own personal iPhone + insta360 go collection method. Once I have a diverse variety of datasets, I'll double down on what I believe to be the key algorithms required to make useful data for imitation learning 📊 1. Camera Pose estimation via SLAM/SFM for ego perspective (and automatic calibration for exo) 2. Human pose estimation for both egocentric + exocentric views 3. Metric 3D reconstruction + object tracking I'll be setting up reasonable open-source baselines for each of these to validate that these datasets work, and then finally try to use the generated datasets for some imitation learning via the pi0-lerobot repo I've been working on. I plan on making a blog post + providing more info on all of this in the near future so stay tunedshow more

A Berkeley chemical engineer who suffered years of unexplained... illness discovered through a spy camera from Amazon that her coworker, David Xu, was poisoning her food and water with cadmium, a toxic heavy metal for nearly 18 months between 2017 and 2019. The victim experienced significant health problems and at times required emergency care at a hospital. Newly released video shows Xu appearing to put a substance in her water bottle in 2019. Xu was arrested that year and charged with attempted murder and two counts of poisoning, offenses that could have sent him to prison for life. But under a plea deal, the attempted murder charge and one poisoning count were dropped. Xu pleaded no contest to one felony poisoning charge and received two years of probation, with credit for one day served. He completed probation in March and is now seeking to have his felony conviction dismissed.show more

Spatial AI ( is building large real-world datasets to... teach robots how to navigate the world and complete tasks. Their first open source dataset, SEA (Spatial Everyday Activities), is the largest curated egocentric dataset of people carrying out real tasks, with 10,000 hours of data.show more

I am posting after a Long Time on Twitter,... but its to announce a big change. I have started and we are Collecting Egocentric Data at Scale from India, Covering 300 + Commercial Locations 1500 + Households This is the network and base we have built in just past 2 months, as the Robotics companies, VLMs and World Models increase their requirements on Real World Data collection, Human Loops will be keep scaling out capacity. We are on track to collect 1M Hours of Egocentric in the coming 6 months for our clients exclusively. We are maintaining 95% Quality standards across industrial data and running a End - End Operational Management complying all Indian Laws and compensating our partners/operators. (From Environment sourcing, to hardware, Legal contracts, deployment, training, collection, processing) Check our Samples: - Commercial Videos -- Household Videos --- Multimodel ---- Egocentric + Live Audio Narration Links : On the Journey to become #1 India Physical AI Data Partner. We are also building our capacity as annotation and labelling partner for data companies to become end-end partner for companies. A big change from the world of crypto and web3 but physical AI and data space is where i want to build my next venture #Egocentric #EgocentricIndia #PhysicalAI #Robotics #India #AIdata #data #Multimodeldata #worldmodels #VLMs #Humanloops #Egocentridata #Multimodeldatashow more

What does a real humanoid robotics dataset actually look... like? Humanoid robots are expensive and complex, which makes their data unusually difficult to collect - and difficult for most people to get hands-on experience with. We previously shared a post on conversion code for the new Human in the Wild dataset by BitRobot 🦾, Hugging Face, and Unitree. Now we’re publishing the full HIW-500 data converted to RRD, so you can explore that data directly with Rerun. Watch the robot move. Inspect cameras and robot states together. Zoom in on a moment and query episodes through Rerun catalog! If you’re curious about a real-world humanoid dataset, HIW-500 is now a lot easier to explore. We share the conversion code along with our previous abc-130k dataset conversion example.show more

I’m honored to share that I have been appointed... the first Global Ambassador for WR Chess This new chapter will take me across continents, promoting chess and impacting lives on a global stage from Africa, Asia, Latin America tour to a world record event in Peru in June. The CEO of WR Chess Wadim Rosenstein has quietly supported the African chess community for four years. Together I believe we can achieve so much more for the chess world. We agreed to meet in Paris and travel to Prague for the chess festival. And for hours, we spoke about a global vision and a legacy for chess development and industrialization across Africa and beyond. One important thing Wadim and I share in common is we both came from nothing but chess opened doors we never imagined. Now we want to extend this to the world. I am proud to join the WR Chess family. Even more excited for what we will build together. It is possible to do great things from a small place.show more

Eastworlds aims to build the largest dataset of high-quality... humanoid teleop data outside of China. While we believe human egocentric data will play a big role in pretraining embodied AI, robot teleop data remains the closest thing to expert demonstration.show more

On 20th of April 1976, R.B. Ramesh was born!... 50 years ago! He became a GM at the age of 27 years in 2003. In 2002, he had won the British Championships, and in 2007 he became the Commonwealth Champion. In 2008, he took a bold step in his chess career. He gave up playing chess and started a chess academy and named it Chess Gurukul. He also gave up his job in Indian Oil (IOCL) around the same point. Many told him that this was a dangerous decision. But Ramesh had 2 things going for him: 1. He loved coaching 2. His better half WGM Aarthie Ramaswamy was also ready to support him in this endeavour! With his immense chess knowledge, willingness to learn and ability to work extremely hard, Ramesh created the base for a real chess boom of Indian chess! Players like Praggnanandhaa, Vaishali, Aravindh Chithambaram, Karthikeyan Murali, Bharat Subramaniyam and many others grew up at his academy! There was a point when almost every single rising talent of Indian chess wanted to work with him. Today as Ramesh celebrates his 50th birthday, it is interesting to see the road he has travelled. Vaishali is going to play the World Championship Match in a few days! He also received the Dronacharya award recently. He started a completely free chess academy named Chola Chess to develop the next generation of Indian chess. Thank you Ramesh for powering Indian chess! Your contribution to the growth of the sport in the country has been immense. Wishing you a very happy 50th birthday. Photo: Tushar Damor #chess #chessbaseindia #rameshchessshow more

So much of the complexity for multimodal/robotics data is... just getting the right format that's both flexible and performant. I dig into it here in the context of 3d reconstruction. I'm hoping to build an opensource arkit like pipeline and explain the first bit for the data pipeline here! Heres a video of what the final data looks like in a registered catalogshow more

Trained on zero real-world data. Learned to walk, pick... up boxes, and follow multi-step instructions... in the REAL world. ( 📌 Paper below) Researchers from Amazon FAR, Berkeley, Stanford, and CMU scanned real rooms with an iPhone, rebuilt them as 3D Gaussian Splatting scenes, then generated 48,000 synthetic trajectories of a Unitree G1 walking, grasping, and placing objects inside those virtual replicas. They rendered the robot's first-person camera view from each run and paired it with the matching language instruction and motion data. That's the dataset every humanoid team needs and nobody has: synced egocentric video + language + kinematics, at scale. Instead of collecting it in the real world, they manufactured it. They trained a vision-language-kinematics policy on that synthetic data alone, then deployed it on the physical G1 across five task types: navigation to a named object, lifting boxes of three different sizes with no per-size tuning, chained multi-step tasks, robustness to mid-task layout changes and flickering lights, and multi-minute long-horizon runs. No real-world fine-tuning at any point. Real-world interaction data has been the hard limit on humanoid learning... slow, expensive, and small. If scanning a room once and synthesizing thousands of labeled interactions holds up as a general recipe, that limit moves. Data stops being the bottleneck robotics teams have to solve for. 📌 Paper: Project: ——- Weekly robotics and AI insights. Subscribe free:show more

Suno got hacked in November 2025. That same month,... they raised $250 million. They told zero customers. Over the next eight months, they raised another $400 million on top of that. Their valuation more than doubled to $5.4 billion. Hundreds of thousands of users had their emails, phone numbers, and Stripe payment data in a hacker's hands the entire time, and Suno said nothing. The leaked source code confirmed what the music industry has been saying in court. Suno scraped 2 million YouTube music clips, 62,000 hours from Pond5, 12,000 hours from Deezer, and planned to grab a million hours of podcasts. They used Bright Data proxy rotation to bypass YouTube's anti-bot protections. The scraping instructions are in the actual codebase, logged by platform. When 404 Media broke the story, Suno called it a "limited security incident" and said individual notifications "were not warranted." They scraped human art without asking. They got breached and hid it for eight months while raising $650 million. The artists, the users, and the investors all got the same treatment. Suno took what it needed from each of them and moved on. That's what an AI company looks like when it sees the world as a training set.show more

🇷🇺🇮🇳 Russia’s Gold Exports to India Hit 5-Year High... – Report Russia exported over $50mn in gold to India from January to October last year, RIA Novosti reported, citing data from the 🇮🇳 Ministry of Commerce and Industry. Deliveries resumed in August with a $23.4mn shipment, the largest since 2021 – with a further $26.9mn sent in October. India remains one of the world’s biggest gold importers, with demand up 16.5% to $50.7bn in the first 10 months of 2025.show more

I’ve been using Rork as a partner for over... a year and have built more than 10 projects with it, including both apps and games. You can absolutely build an app from development all the way to a production launch on the App Store with AI. Users don’t care what you used to build your app. They care whether it solves their problem and how well you distribute it. For an MVP to test the market, you don’t need a huge team, an outsourcing company, or to spend tens of thousands of dollars. 🥴show more