Introducing GridSFM, a small foundation model that can predict... AC optimal power flow in milliseconds, boosting efficiency and unlocking cost savings. Learn how GridSFM gives grid operators direct visibility into congestion, stability, and system health:show more

Production-ready RAG systems need observability. From tracking latency and... throughput to evaluating response quality with human feedback or LLM-as-a-judge, robust observability gives you visibility into both system performance and output quality, on both a component and system-wide level. This lesson from our Retrieval Augmented Generation course breaks down the core components of an effective eval system and how to balance cost, automation, and accuracy when choosing your metrics. 📚 Learn more in the full course:show more

Jerry Stackhouse walks through a secondary break concept rooted... in Dean Smith’s North Carolina system and explains how it can flow into more modern actions with the same alignment that can evolve into five-out spacing.show more

Humans learn and improve from failures. Similarly, foundation models... adapt based on human feedback. Can we leverage this failure understanding to enhance robotics systems that use foundation models? Introducing AHA—a vision-language model for detecting and reasoning over failures in robotic manipulation. Project page: 🧵Thread👇 Aha!show more

Liquid AI (Liquid AI) CEO Ramin Hasani (Ramin) says:... "We're bringing the cost of tokens to zero." "The axis was maximizing intelligence at all costs." Foundation models need to optimize across 3 axes: 1.) Intelligence and capability 2.) Efficiency and cost 3.) Substrate: where the intelligence actually runs "Efficiency is not an afterthought. Energy is not abundant." "If you maximize intelligence at all costs, that axis alone is not going to get you to the place that you want to go." "Efficiency and cost of intelligence as a first-class citizen, and not an afterthought." "Where does this intelligence system go? AI is majorly getting hosted in data centers, but you could also bring intelligence on phones, on laptops, on airplanes, on cars." "Imagine if it's not tokens that are actually important. It's just the outcome that actually matters." "Not just thinking about foundation models as the token machines that are generating money and revenue for the foundation model companies that are useless tokens, and getting them into the place where they can actually unlock true value for enterprises."show more

The first integrated energy interconnection system in Beijing Daxing... Airport City (BDAC) is now officially up and running! Here at the energy center serving both the comprehensive bonded zone of Beijing Daxing International Airport and the surrounding non-bonded areas, the system functions as an "energy super brain" powered by big data, coordinating heating, solar power, and EV charging across the region. Inside the dispatching center, a huge digital screen acts like the central nervous system, monitoring energy flows and making real-time adjustments to ensure efficiency and stability. The launch of this system not only improves efficiency and saves energy, but also delivers a green smart energy solution that can be replicated and promoted elsewhere. It is injecting steady, sustainable power into the high-quality development of BDAC.show more

Today we’re launching CredAgent, the most transformational product ever... built for credentialing. An autonomous agent that can reason, verify, and execute complex credentialing workflows end-to-end, with human oversight. Credentialing today still relies on outsourced BPOs and manual workflows. This isn’t another workflow tool or co-pilot. It’s a fundamental shift in how credentialing work gets done. CredAgent delivers significant cost savings, better control, and higher quality and early pilots are already showing up to 10× efficiency gains. In a world of autonomous agents, APIs are foundational. We’ve been API-first since day one, and remain the only truly API-first credentialing company , uniquely positioned to power real healthcare agents. If you’re a health plan, virtual care network, or health system ready to modernize credentialing, I’d love to show you CredAgent.show more

Introducing OpenVPP's Routing Service for Grid Operators. The Routing... Service is an off-chain connectivity service with standardized APIs that allow for 3rd parties (device aggregators, utilities, or program administrators) to gather the necessary information to form a DER connection. A program enrollment page calls the Routing Service API and provides the following standardized information: -->Site address -->Utility -->Device make and model -->Device identifier (e.g. serial number) -->Desired connectivity protocol (e.g. OpenADR 2.0k) -->Desired program identifier A fee is assessed in $OVPP for connecting the device to the utility, and can be paid for in one of three ways: 1) Direct payment in native $OVPP from the customer’s wallet. 2) Indirect payment in native $OVPP from the program administrator’s wallet. 3) Direct or indirect payment in stablecoins via OpenVPP’s Stablecoin On-Ramp Service, which automatically purchases the necessary $OVPP for the transaction. Learn more on about OpenVPP's Routing Service for Grid Operators - The Internet of Energy.show more

PG&E and Tesla have just announced that the Cybertruck,... Tesla Powershare Gateway and Tesla Universal Wall Connector have all been approved for participation in PG&E's residential Vehicle‑to‑Everything (V2X) program, marking a major milestone for bidirectional EV integration in California. This marks the first AC vehicle-to-grid application approved for customers in California, enabling vehicles to connect using much simpler equipment rather than specialized DC infrastructure. PG&E customers with Cybertrucks will now be able to use their vehicle to power their home in an outage and earn money by selling power back to the electric grid during grid events. "Electric vehicles can do more than move people—they can help power homes and support the grid," said Jason Glickman, Executive VP of Strategy and Growth, PG&E. "By welcoming Tesla into our residential V2X program, we're expanding customer choice while making California's grid more flexible, resilient, and affordable. Powershare Grid Support enables Tesla vehicles to strengthen our electricity system, while earning money for EV owners." More info below:show more

When Digital Public Infrastructure (DPI) meets leadership, inclusion scales.... In Rwanda, Rwanda Information Society Authority and Sharon Umunyana are deploying Mojaloop Foundation to build an inclusive instant payments system that cuts costs, retains national ownership and keeps data secure. As we head into the Global DPI Summit 2025 2025, it is clear that with open, interoperable DPI, governments can connect citizens to safe, affordable financial services, empowering women, small businesses and families to fully participate in the digital economy. Learn more about Sharon's story and Mojaloop's impact, powered by the Gates Foundation: #InclusiveGrowth #GlobalDPISummitshow more

In 2024, the world added nearly 600 gigawatts of... new renewable power — the largest annual increase in history, driven overwhelmingly by solar and wind. Our new government is matching that ambition here at home. From world-class wind resources in Atlantic Canada to vast solar potential across the Prairies, Canada is poised to generate new clean, cost-competitive electricity. Tune into my latest video to learn more about the opportunities ahead and how a cleaner, more reliable, and more sustainable grid is within reach 👇🎥show more

SOMEONE BUILT A MAP THAT SHOWS EXACTLY WHERE EVERY... POWER PLANT, TRANSMISSION LINE, SUBSTATION & DATA CENTER SITS ON THE US GRID all on one interactive map. all free you can see how the grid is laid out... where the datacenters cluster... which transmission corridors carry the load... where the high-capacity connection points are zoom into any region and the whole picture comes into focus why energy costs what it costs, why data centers go where they go, why some states are power exporters and others aren't this is the kind of infrastructure visibility that used to require expensive industry reports now it's one tabshow more

Yann LeCun (Yann LeCun) just revealed why he left... Meta, Why are LLMs an extremely narrow field, and how "World model" is the way to build really meaningful agentic future. 🎯 Beautiful and simple in 1.5 mints. "I can’t imagine we can build agentic systems without those systems having the ability to predict, in advance, what the consequences of their actions are going to be. The way we act in the world is that we can predict the consequences of our actions, and that’s what allows us to plan. So what is a world model? Given the state of an environment, a system you want to control at time t, and given an action or intervention you imagine taking, can you predict the state of the world (or the system) at time t + 1? If you can, that’s a world model. You don’t do this at a pixel level, if it’s video. You do this in an abstract representation space, and that’s a crucial key insight." --- From 'AI House Davos" YT channel (full link in comment)show more

Why are major financial institutions looking to Avalanche's Subnet... architecture to test how blockchains and tokenization could revolutionize capital markets? Ada Kokoshi, Digital Assets Director DTCC, explains one key advantage that her team saw as a major benefit to using Avalanche Evergreen subnets. Speed. When trying to facilitate atomic settlement with high-value transactions, sub-second finality can be essential in bringing efficiency and cost-savings to private markets.show more

In continuous cable manufacturing, compressed air performance must keep... pace with constantly changing production demands. In a leading cable manufacturing facility, compressed air plays a critical role across extrusion controls, air-wipe drying, cleaning, and material handling systems, making pressure stability and compressor reliability essential for consistent product quality and high uptime. At the plant, a 75 kW fixed-speed compressor reliably met base air requirements. However, variations in production loads, line speeds, and process cycles led to frequent demand fluctuations, resulting in repeated compressor cycling, higher energy consumption, and pressure instability across the system. With the deployment of ELGi’s DEMAND=MATCH System, performance was evaluated under actual operating conditions by comparing operations with and without DEMAND=MATCH. The results were clearly validated. • 11.1% reduction in average power consumption (from 69.9 kWh to 62.1 kWh per hour) • Approximately 1,489 units of energy saved over just eight days of actual running • Reduced load cycling, leading to improved pressure stability and lower mechanical stress on the compressor This deployment demonstrates how intelligent airflow control can deliver measurable gains in energy efficiency, system stability, and operational reliability, all critical in a continuous manufacturing environment. A strong example of how aligning compressed air delivery with real-time demand can drive sustainable performance improvements on the shopfloor. Learn more about how DEMAND=MATCH optimises compressed air systems:show more

The traditional approach to group healthcare coverage forces employees... into “one-size-fits-none” options. But with ICHRA (Individual Coverage Health Reimbursement Arrangement), employers can finally offer real personalization at scale. Louis DeStefano (SVP, Growth at Oscar Health) and Dr. Jon Van Der Veer (Founder & CEO, Hy-Vee Health Exemplar Care), talk about how ICHRA transforms the way that employers offer healthcare benefits. Here’s how ICHRA works: Employers contribute a predetermined amount of money to a flexible account for each employee. Employees get to choose from a wide range of individual plans that best fit their needs. With this model, employees get the flexibility they need while employers get greater cost control and predictability. Learn how Oscar is bringing true choice to employee benefits atshow more

I’ve had so many health conversations on this show... and it’s almost at a point where I’m wondering if there’s much more that I’ve got to learn. And then someone like Dr Rhonda Patrick sits across from you… and I realised how wrong I was. This is my second time speaking to Rhonda and after our first chat, I knew we had to speak again. Dr Rhonda is a world-leading biomedical scientist who has dedicated her life to uncovering the truth about how to slow down aging and achieve optimal health. She stays at the very cutting edge of the studies that are coming out and she’s so good at articulating them in a simple way that someone like me can understand. It means that I can live a happier, healthier life and so can you.show more

We often hear that machine learning models "learn patterns... in data". But what does that actually look like in geometry? If you dropped a little elastic mesh into a cloud of points and let it learn, how would it fold itself to match the shape of the data? In this scene we watch a self-organizing map...a simple unsupervised neural model...learn the shape of a two-dimensional dataset arranged in a spiral arm. On top of this, we lay down a square grid of neurons whose weights live in the same plane. At the start, this grid is just a flat net floating across the cloud...it knows nothing about the structure underneath. Learning is a repeated game: pick a random data point, find the neuron whose weight is closest, and then nudge that neuron and its neighbours toward the point. Do this again and again, while slowly shrinking how far the neighbourhood influence spreads. #MachineLearning #ManifoldLearning #UnsupervisedLearning #NeuralMaps #GeometricMLshow more

We often hear that Machine Learning models learn patterns... in data. But what does that look like in geometry? Picture dropping an elastic mesh into a cloud of points and letting it adapt. How would it bend, stretch, and settle so it matches the shape hiding in the data? In this scene we watch a self-organizing map(SOM)...a classic unsupervised neural model...learn a 2D dataset arranged like a spiral arm. Over the points we place a square grid of neurons whose weights live in the same plane. At the start it’s just a flat net drifting across the cloud. No clue what the structure is. Fir the SOM, learning is a repeated game: Pick a random data point, find the neuron whose weight is closest, then nudge that neuron and its neighbours toward the point. Repeat, repeat, repeat...while gradually shrinking how wide the neighbourhood influence spreads. The result is satisfying...the grid stops being a grid and turns into a coordinate sheet wrapped onto the spiral. #MachineLearning #ManifoldLearning #UnsupervisedLearning #NeuralMaps #GeometricML #SelfOrganizingMap #Topologyshow more

Welcome to the Lab of the Future! 🧬🤖 Excited... to share LUMI-lab, out today in Cell — a self-driving platform that pairs an AI foundation model with a robotic lab to autonomously discover ionizable lipids (LNPs) for mRNA delivery. The core problem: Designing lipid nanoparticles (LNPs) is hard. The chemical space of ionizable lipids is vast, experimental cycles are slow, and — critically — historical LNP datasets are far too small to train a predictive model from scratch. Most AI approaches in this space hit a wall immediately: not enough data to learn from. Our solution: lab-in-the-loop foundation model learning. Instead of training on LNP data alone, LUMI starts as a transformer-based foundation model pretrained across broad chemical space, building rich molecular representations before it ever sees a single LNP experiment. Then it enters a closed loop with a robotic synthesis platform: predict → synthesize → assay → update. Each round of real wet-lab experiments fine-tunes the model, which then proposes smarter candidates for the next round. The lab isn't just validating AI predictions — it's actively teaching the model, continuously. What happened when we let it run: LUMI-lab autonomously synthesized and screened 1,700+ ionizable lipids in human bronchial epithelial cells. The top candidate — LUMI-6 — features a brominated lipid tail, a structural motif that had been largely overlooked in LNP design. LUMI found it without being told where to look. When formulated into LNPs and delivered intratracheally to mice, LUMI-6 achieved 20.3% gene editing efficiency in lung epithelial cells — a compelling result for one of the hardest-to-reach therapeutic targets, directly relevant to diseases like cystic fibrosis and alpha-1 antitrypsin deficiency. Why this matters beyond LNPs: This is a proof of concept for a broader thesis — that foundation model pretraining + active learning + robotic experimentation can overcome the data scarcity bottleneck that plagues AI-driven discovery in biology. You don't need a massive domain-specific dataset to start. You need a model that can generalize, a lab that can generate the right data, and a loop that connects them. Huge congratulations to first authors Yue Xu, Haotian Cui, and Kuan Pang, and to the entire Bowen LI team. Grateful to our collaborators at University Health Network and Leslie Dan Faculty of Pharmacy, and to Princess Margaret Cancer Centre Research Princess Margaret Cancer Centre Research. 📄 Paper:show more