AI In Scientific Research

🧠 This AI Glider is Mapping the Ocean 100x Faster Than Humans Flying Fish Technologies Pty Ltd are transforming marine monitoring. Their AI-powered underwater gliders are capturing the ocean like never before: 🚤 100x faster than traditional methods 📍 15+ geotagged data points per second 📸 6 million images analysed by machine learning 🐠 Mapping everything from fish to fragile benthic habitats in clear 3D photogrammetry Recent highlights from their mission to the Red Sea: → 350km of continuous reef surveyed → 3.5M images captured in under a month → 200M datapoints generated in just 2 days They provide real-time, high-resolution, high-impact intelligence, powering decisions for ocean conservation and climate resilience, enabling: ⤷ Photorealistic digital twins to track change over time ⤷ AI-driven habitat classification and species detection ⤷ Driverless, boat-based gliders that follow terrain and depth This is the kind of NatureTech that moves us from scattered data to smart, systemic ocean protection. Excited to follow David Kettle and the team at FFT’s progress on this! #OceanTech #MarineScience #NatureTech #AIforNature #BlueCarbon

𝗔𝗜 𝗳𝗼𝗿 𝗚𝗢𝗢𝗗: 𝗡𝗔𝗦𝗔 𝗮𝗻𝗱 𝗜𝗕𝗠 𝗹𝗮𝘂𝗻𝗰𝗵 𝗼𝗽𝗲𝗻-𝘀𝗼𝘂𝗿𝗰𝗲 𝗔𝗜 𝗳𝗼𝘂𝗻𝗱𝗮𝘁𝗶𝗼𝗻 𝗺𝗼𝗱𝗲𝗹 𝗳𝗼𝗿 𝗺𝗼𝗿𝗲 𝗲𝗳𝗳𝗶𝗰𝗶𝗲𝗻𝘁 𝘄𝗲𝗮𝘁𝗵𝗲𝗿 𝗮𝗻𝗱 𝗰𝗹𝗶𝗺𝗮𝘁𝗲 𝗳𝗼𝗿𝗲𝗰𝗮𝘀𝘁𝗶𝗻𝗴! 🌍 (𝗧𝗵𝗶𝘀 𝗶𝘀 𝘄𝗵𝗮𝘁 𝘀𝗵𝗼𝘂𝗹𝗱 𝗴𝗲𝘁 𝗺𝗼𝗿𝗲 𝘀𝗽𝗼𝘁𝗹𝗶𝗴𝗵𝘁 𝗽𝗹𝗲𝗮𝘀𝗲 𝗮𝗻𝗱 𝗡𝗢𝗧 𝘁𝗵𝗲 𝗻𝗲𝘅𝘁 𝗖𝗵𝗮𝘁𝗚𝗣𝗧 𝗪𝗿𝗮𝗽𝗽𝗲𝗿!) In collaboration with NASA, IBM just launched Prithvi WxC an open-source, general-purpose AI model for weather and climate-related applications. And the truly remarkable part is that this model can run on a desktop computer. 𝗛𝗲𝗿𝗲'𝘀 𝘄𝗵𝗮𝘁 𝘆𝗼𝘂 𝗻𝗲𝗲𝗱 𝘁𝗼 𝗸𝗻𝗼𝘄: ⬇️ → The Prithvi WxC model (2.3-billion parameter) can create six-hour-ahead forecasts as a “zero-shot” skill – meaning it requires no tuning and runs on readily available data. → This AI model is designed to be customized for a variety of weather applications, from predicting local rainfall to tracking hurricanes or improving global climate simulations. → The model was trained using 40 years of NASA’s MERRA-2 data and can now be quickly tuned for specific use cases. And unlike traditional climate models that require massive supercomputers, this one operates on a desktop. Uniqueness lies in the ability to generalize from a small, high-quality sample of weather data to entire global forecasts. → This AI-powered model outperforms traditional numerical weather prediction methods in both accuracy and speed, producing global forecasts up to 10 days in advance within minutes instead of hours. → This model has immense potential for various applications, from downscaling high-resolution climate data to improving hurricane forecasts and capturing gravity waves. It could also help estimate the extent of past floods, forecast hurricanes, and infer the intensity of past wildfires from burn scars. It will be exciting to see what downstream apps, use cases, and potential applications emerge. What’s clear is that this AI foundation model joins a growing family of open-source tools designed to make NASA’s vast collection of satellite, geospatial, and Earth observational data faster and easier to analyze. With decades of observations, NASA holds a wealth of data, but its accessibility has been limited — until recently. This model is a big step toward democratizing data and making it more accessible to all. 𝗔𝗻𝗱 𝘁𝗵𝘀 𝗶𝘀 𝘆𝗲𝘁 𝗮𝗻𝗼𝘁𝗵𝗲𝗿 𝗽𝗿𝗼𝗼𝗳 𝘁𝗵𝗮𝘁 𝘁𝗵𝗲 𝗳𝘂𝘁𝘂𝗿𝗲 𝗼𝗳 𝗔𝗜 𝗶𝘀 𝗼𝗽𝗲𝗻, 𝗱𝗲𝗰𝗲𝗻𝘁𝗿𝗮𝗹𝗶𝘇𝗲𝗱, 𝗮𝗻𝗱 𝗿𝘂𝗻𝗻𝗶𝗻𝗴 𝗮𝘁 𝘁𝗵𝗲 𝗲𝗱𝗴𝗲. 🌍 🔗 Resources: Download the models from the Hugging Face repository: https://lnkd.in/gp2zmkSq Blog post: https://ibm.co/3TDul9a Research paper: https://ibm.co/3TAILXG #AI #ClimateScience #WeatherForecasting #OpenSource #NASA #IBMResearch

Google DeepMind created a Gen AI model to predict extreme heat, and cyclones -- and it's faster and more accurate than traditional prediction models. It's going to be a huge deal as the climate crisis keeps getting worse. The model's called GenCast, and it uses a diffusion model, similar to those in image generation, adapted for Earth's spherical geometry. The model was trained on four decades of weather data from ECMWF's ERA5 archive. It generates 50+ possible weather scenarios, giving probabilistic ensemble forecasts. These forecasts predict daily weather and extreme events like cyclones with high accuracy. GenCast operates faster and more efficiently than traditional systems, needing just 8 minutes per forecast using TPUs. GenCast outperformed ECMWF’s ENS on 97.2% of forecasting targets, especially for extreme heat, wind, and cyclones. Its speed and precision help safeguard lives, improve renewable energy reliability, and support climate resilience. #GenAI #AI

AI x Quantum is the new the frontier of materials innovation. The latest data on the hottest, nascent manufacturing markets highlights the companies driving breakthroughs in materials development; fueled by $1.1B in funding this year. What's driving the surge? AI and quantum computing advances are helping these platforms reach accuracy levels that can replace physical trials at a fraction of the cost and time. Key developments at the intersection of quantum, AI, and materials: ↳Full-stack integration: Radical AI combines AI, quantum mechanics, and automated chemical characterization in a single platform ↳Data management revolution: Uncountable Inc. handles experimental data collection, management, and visualization – letting researchers focus on discoveries instead of searching for data ↳Proven cost reduction: Kebotix combines machine learning with lab automation, reporting 5x reduction in lab costs ↳Quantum algorithms advancing: Qunova Computing claims their algorithms reduce computational requirements by over 1,000x compared to traditional methods ↳Infrastructure scaling: Companies like Albert Invent combine material development with laboratory information management and regulatory compliance The shift from tools to platforms is critical. These aren't just simulation tools – they're comprehensive R&D management systems positioned to replace entire suites of disparate research software. Now, we're seeing a new breed of materials companies founded by AI experts, exemplified by former OpenAI researcher Liam Fedus's Periodic Labs that just raised $200M at a $1B valuation. When AI's top talent moves into materials, it signals where the next wave of industrial innovation will emerge. SandboxAQ, the Google spinout with the highest Mosaic score (879), develops AI and quantum models to predict molecular properties. While fault-tolerant quantum computers aren't expected until 2030, companies like Quemix Inc. are developing quantum-inspired techniques providing advantages today. QpiAI notably built its own quantum computer using superconducting circuits. Broadly, these emerging platforms transform materials development from years to months by digitally testing compounds before expensive lab work begins; shifting the competitive edge from lab size to the rapidly-expanding computational limits of intelligence. P.S. Want more insights on the companies developing the future of materials? Drop "material developments" in the comments for *free* access to CB Insights' data and insights on the Material development platforms market.

When we talk about the power of AI, we often focus on its business benefits. But its most profound impact lies in how it helps us solve complex, real-world problems - including helping us understand and protect our natural world. A great example of this is happening right now off the coast of Australia. Our Google Research team—led by Data Scientist Lauren Harrell—headed out to the Sunshine Coast alongside our partners at Griffith University to listen to our oceans. Sound is critical underwater. It is the primary way marine mammals communicate, navigate, and survive. But listening to the ocean means processing thousands of hours of audio. To help scientists, we are using AI to analyze audio captured by specialized underwater microphones (hydrophones). AI automatically scans these massive recordings to identify the complex, beautiful songs of migrating humpback whales. By instantly translating this noise into clear tracking data, we can help scientists map migration paths in real-time. This is critical for predicting shipping lanes, preventing ship collisions, and understanding how climate change and ocean noise are impacting these gentle giants. I’m proud that this work is happening right here in Asia-Pacific. Through Google’s Digital Future Initiative, this underwater microphone network will soon span a massive 2,000-kilometer stretch of the Australian coastline, and I'm excited to see how else AI can solve our region's biggest challenges. Photo credits: Ste Everington (@steunderwater) Johnny Gaskell (@johnny_gaskell)

Forest carbon monitoring gets an AI boost, reports Abhishyant Kidangoor. Forests have long been surveyed from above. Satellite data reveal where they stand and how they shrink or grow, while lidar—laser-based radar—has allowed scientists to map them in 3D, uncovering details that lie beyond human sight. Now, artificial intelligence is adding a new layer of insight. Earth-imaging company Planet has unveiled a Forest Carbon Monitoring tool that fuses its satellite imagery with lidar data. The tool can estimate carbon storage, tree height, and canopy cover in remote forests at a granular resolution of three meters. “It will help us understand aspects of the forest that might not be initially accessible to the naked eye,” says Andrew Zolli, Planet’s chief impact officer. Satellites track forest cover but not the carbon stored in biomass. Measuring this requires lidar, which calculates tree dimensions by measuring the time laser beams take to bounce off foliage. NASA’s GEDI mission, mounted on the International Space Station, has mapped swathes of forests, but coverage gaps persist. Planet’s tool aims to bridge these voids, training machine-learning models to infer carbon data in areas without lidar coverage. Initial findings from the tool have been striking. While deforestation ravages the Amazon, the northern reaches harbor untouched carbon reserves. “What really resonated with me is the understanding of where we have extant forest carbon stocks which we must absolutely protect,” says Zolli. The data also underpin Project Centinela, which supports conservation efforts in biodiversity hotspots like Tanzania’s Gombe Stream National Park. Meanwhile, carbon markets—often criticized for opacity—may gain credibility through applications of the tool argues Zolli: “The data gives a shared, common picture of what’s actually happening on the ground.” Planet’s innovation rests on decades of data, cutting-edge AI, and cloud computing. “We are the first generation that has had all three in place,” Zolli says, enabling swift, confident assessments of carbon across the globe. 📰 story: https://lnkd.in/gwRWf5Qf 📷: A view of carbon storage in forest and an area of fishbone deforestation in the Brazilian Amazon. Image courtesy of Planet.

In the past few years, I have worked quite a lot on GreenTech and climate AI and have shared resources on the same. Today is one such day again!   Chile's Nahuelbuta mountain range is an awe-inspiring tapestry of biodiversity, teeming with unique species such as the Darwin's fox. However, this ecosystem constantly faces threats from human activities, wildfires, and encroachment. With less than 1,000 Darwin's foxes remaining, their existence is hanging by a thread. Enter the "Nature Guardian" initiative – a collaborative marvel involving Rainforest Connection (RFCx), deploying solar-powered devices enriched with AI to monitor and safeguard this invaluable ecosystem vigilantly. "Nature Guardian" is supported by Huawei’s #TECH4ALL, which is always committed to enable an inclusive and sustainable digital world. These ingenious devices have evolved into the unseen sentinels of this diverse landscape, ceaselessly engaged in environmental monitoring, tracking animal calls, and swiftly identifying threats like illegal logging and poaching. Meticulously positioned high in the treetops, they provide round-the-clock coverage, seamlessly linked to a cloud-based AI platform. One of the project's noteworthy facets lies in its AI analytics, expertly trained to recognize various animal species. This empowers researchers to scrutinize their distribution and behaviours, offering invaluable insights for adaptive conservation measures. What I also found interesting is the system's ability to issue real-time alerts via a mobile app if any threat is detected, enabling rapid responses to protect this delicate ecosystem. As of August 2021, five Nature Guardian devices and ten edge devices had been deployed, covering 30 km2 of Nahuelbuta forest. However, the project's vision doesn't halt here; it's expanding to Chiloe Island and the coastal regions of the Valdivian forest, where sightings of the elusive Darwin's fox have been reported. This endeavour underscores the power of collaboration among organisations like RFCx, Bioforest, Etica en los Bosques, the Ministry of the Environment for Chile, and Huawei. In an era where climate change and forest degradation pose significant challenges to ecosystems worldwide, their combined expertise in conservation and technology is contributing towards preserving Chile's unique biodiversity. #innovation #technology #artificialintelligence #greentech #techforgood

Researchers at Stanford University are revolutionizing how we understand the massive Antarctic ice sheet and its potential impact on global sea levels. Using advanced machine learning techniques, they've revealed hidden physics of ice movements that could reshape climate change predictions. 🌊📉 Key points: 1) Massive impact: Antarctica's ice holds enough water to raise sea levels by a staggering 190 feet! Accurate predictions are crucial for preparing coastal areas worldwide. 🏝️ 2) Machine Learning magic: By analyzing satellite and radar data, researchers are using AI to dive deeper into how ice moves and melts. This approach offers insights that traditional models have missed. 🚀 3) Revealing complexity: Most ice models assume uniform properties, but this study shows that ice behaves differently in various directions. Imagine trying to cut a log along its grain vs. against it. ✂️ 4) Future predictions: With better understanding, these advanced models can help predict how Antarctic ice will evolve as the planet warms, influencing sea-level rise predictions and more! 🔮 #AntarcticIce #AIResearch #ClimateChange #MachineLearning #EarthScience #SeaLevelRise #StanfordUniversity

From plots to pixels - how UAV-LiDAR is revolutionising forest inventory in New Zealand The future of forest inventory is airborne-and it’s arriving faster than ever. A groundbreaking national study across radiata pine trials in New Zealand has shown that UAV-mounted LiDAR, combined with machine learning, can accurately predict individual tree diameter and volume with minimal fieldwork. This marks a major shift from traditional, labour-intensive methods to scalable, data-driven forest management. By training random forest models on LiDAR-derived canopy metrics, researchers demonstrated that even with a reduced number of on-ground measurements, prediction accuracy remained high. This opens the door to faster, more cost-effective forest inventories - without compromising precision. Why does this matter? Radiata pine dominates New Zealand’s plantation forestry sector, which is vital for timber production, exports, and carbon sequestration. Efficient monitoring of these forests is essential for sustainability, climate resilience, and economic performance. This research brings us closer to a generalised, low-cost inventory model - one that could transform forest management not just in New Zealand, but globally. Michael Watt I Sadeepa J. I Mikey Mohan, PhD I Robin Hartley I Nicolò Camarretta I Ben Steer I Weichen Zhang I Mitch Bryson I Scion I Ecoresolve I University of Sydney #Forestry #Innovation #UAVLidar #RadiataPine #MachineLearning #ForestInventory #SustainableForestry #RemoteSensing #CarbonSequestration #NewZealand #AI #PrecisionForestry #EnvironmentalTech #Bioeconomy https://lnkd.in/g5QhHdQy

How do we build AI for science? Augment with AI or replace with AI? Popular prescription is to augment AI into existing workflows rather than replace them, e.g., keep the approximate numerical solver for simulations, and use AI only to correct its errors in every time step. The other extreme is to completely discard the existing workflow and replace it fully with AI. We have seen this approach win in areas like weather forecasting. Such end-to-end AI is significantly better for speed: 1000-million x faster. In our latest paper, we show end-to-end learning also wins in data efficiency, which is counterintuitive. Where do these savings come from? The former approach that augments AI relies only on fully accurate training data that is expensive. But end-to-end learning can use both approximate and accurate training data, if the model can learn how to mix them correctly. In many physical systems, coarse-grid numerical solvers yield approximate data while fine-grid solvers fully resolve the scales and yield exact answers. It turns out that Neural Operators offer a perfect solution when such multi-fidelity and multi-resolution data is available, and can learn with high data efficiency requiring only a small amount of fully resolved data, since it can also utilize approximate training data. In contrast, the standard approach of augmenting AI to a coarse-grid numerical solver (closure model) can only train on fully-resolved simulations, making it very expensive and hard to train. Our results are applicable in multi-scale chaotic systems that have traditionally required running long simulations at high resolution such as climate change or plasma in nuclear fusion and astrophysics. Now you can replace expensive simulation fully with AI (Neural Operators), and also train it without requiring such simulations in large numbers for training in many scenarios.