How AI can both liberate and enslave us.

"The productivity gains from AI tools are undeniable. Development teams are shipping faster, marketing campaigns are launching quicker, and deliverables are more polished than ever. But if you’re a technology leader watching these efficiency improvements, you might want to ask yourself a harder question: Are we building a more capable organization, or are we unintentionally creating a more fragile one?" Check out #Radar article by Brinda Sarathy and Rajeshwari Ganesan to find out: https://bit.ly/4nVVXmP

Are we trading resilience for efficiency as AI reshapes our organizations?In this article, we explore how even well-intended optimization can erode the deeper capacities that make teams adaptable and creative. Proud to coauthor with Rajeshwari Ganesan, whose work on technology continues to inspire. https://lnkd.in/g6xYN3FB

🚀 Harnessing AI to Make Environmental Data Work On 15 October 2025, our CEO Marios Charilaou will give a talk at the Wolfson College Oxford Engineering Society. 📍 The Buttery, Wolfson College, Oxford  🕢 7:30 p.m. GMT  🎙️ Talk Title: Harnessing AI to Make Environmental Data Work: From Field Collection to Strategic Insight About the talk: Environmental data is often fragmented, costly, and inconsistent. In this talk, Marios will explore how AI can reshape environmental consulting — from cutting the costs of field data collection, to improving accuracy, reproducibility, and cross-context comparability of analysis. By combining AI prompt engineering with methods like Qualitative Comparative Analysis (QCA) and fuzzy set calibration, we can turn messy datasets into decision-ready insights. But this is about more than methods: it’s about showing how consulting at the intersection of AI and Knowledge OnDemand can deliver faster, sharper, and more impactful solutions to sustainability challenges. 🌍 This event will be of interest to engineers, scientists, policymakers, and anyone curious about the future of consulting, AI, and environmental strategy. 🍻 And after the discussion, we’ll continue the conversation over pints at the Wolfson bar... Join us! 🔗 Full details here: https://lnkd.in/eYD_98ma #AI #Consulting #KnowledgeOnDemand #EnvironmentalData #WolfsonCollege #Oxford #MorphosisGroup

Conservation is about protecting our planet’s future; AI can make your decisions bolder. Imagine an environmental manager planning a reforestation project: AI models ecosystem impacts, predicts species recovery, and optimizes resource allocation in real time. It’s not about AI saving the planet, it’s about empowering you to make choices that balance science and heart. This is #augmentation, not automation. AI processes vast datasets; soil health, climate patterns, biodiversity metrics, etc., so you can focus on strategy and community impact. According to McKinsey’s 2024 report, “Superagency in the workplace,” organizations using AI for environmental planning cut project costs by 18% and improved outcomes by 22% (McKinsey & Company, 2024). Why? Because AI freed managers to prioritize long-term vision over short-term data slog. A 2024 whitepaper on generative AI echoes this, noting 20% faster conservation project timelines when humans leverage AI insights for decision-making. The danger? Over-relying on AI to dictate plans. That risks sterile solutions that ignore local needs. Instead, use #AI to amplify your human judgment, blending data with empathy for communities and ecosystems. At frontiers.ai Technologies, we believe conservation thrives on human passion; AI just sharpens the strategy. The Earth needs bold stewards. What impact will you drive next?

🤖 ⛵ Great example of how AI is being used to augment expert knowledge: helping to remove abandoned nets in our oceans. 𝗠𝗶𝗰𝗿𝗼𝘀𝗼𝗳𝘁 and 𝗦𝗮𝘁𝘆𝗮 𝗡𝗮𝗱𝗲𝗹𝗹𝗮 did an amazing job getting to the bottom of this in an literal sense! 👉 Deep within the ocean 🌊 , a hidden enemy of all living things is lurking, threatening to destroy marine life by trapping it in an inescapable prison.  👉 Over 500 species including turtles, whales, sharks, and crabs have been impacted by this enemy, known as ghost gear, or abandoned plastic fishing nets that litter our seas and oceans.  👉 Industrial-grade plastics that make up fishing nets are manufactured to survive the harshest conditions.  👉 An unintentional consequence of the toils of commercial fishing, nets are frequently lost—and abandoned—at sea. These nearly invisible traps silently drift through the water like ghosts, continuing to trap and kill marine life.  👉 Removing ghost nets is critical—they take centuries to decompose—but it’s also incredibly difficult to find them. For years, locating ghost nets was mostly a game of luck. Imagine looking for a strand of hair in an 𝗢𝗹𝘆𝗺𝗽𝗶𝗰-𝘀𝗶𝘇𝗲𝗱 𝘀𝘄𝗶𝗺𝗺𝗶𝗻𝗴 𝗽𝗼𝗼𝗹. The water is pitch-black and shifting and the hair is sinking. According to WWF, plastic ghost nets entangle millions of marine animals every year. Cleaning it up may feel overwhelming and impossible, but projects like GhostNetZero.ai are tackling it with AI. Learn more: https://lnkd.in/gqsSZ5QH

How cool is this? 🌿🤖 The H4rmony Project is doing something really unique— combining AI with ecological storytelling to help shift how we talk about sustainability. Working with the University of Gloucestershire, they’re using ecolinguistics (basically, language that supports the planet) to teach AI to tell better, greener stories. It’s all about changing the narrative so tech becomes part of the solution, not the problem. And they’re doing it through communication and storytelling — which makes it not just smart, but very creative!! Definitely worth keeping an eye on 👀 Read more here: https://lnkd.in/ena5Ge2x

“How Physics-Informed AI Is Changing Material Science” A few minutes ago, I read a fascinating piece of news: researchers from KAIST, South Korea have introduced a new method that combines physical laws + machine learning to predict material properties even with very limited data. (Source: Phys.org) This caught my attention because one of the biggest challenges in AI—especially in material science—has always been the lack of large experimental datasets. 🔍 What the article says The KAIST research team developed a new approach called Physics-Informed Machine Learning (PIML), where physical laws are embedded directly into the training process—instead of letting the model “learn only” from data. (Phys.org) Key advantage: Even with limited experimental data, this method allows models to predict material properties more accurately. (Phys.org) This integration makes the model “aware” of physical constraints, so predictions aren’t based only on statistical patterns, but are also grounded in the real-world laws of physics. (Phys.org) 💡 Why this matters for us in AI Adding domain knowledge = more power By embedding principles from physics, chemistry, or other sciences, the model becomes more logical, trustworthy, and reliable—not just a black-box statistical function. A solution for data-scarce domains Complex models usually demand massive datasets, but in areas like material science, generating experimental data is costly and time-consuming. This new method shows that high-quality predictions are still possible with small datasets. Balancing theory and practice True AI power comes when human scientific knowledge and machine learning are combined, rather than relying solely on raw data. Accelerating material innovation If successful, this approach could speed up the discovery of materials with unique properties (for electronics, energy, magnetics, etc.). What once took years could now be achieved in a fraction of the time.  Source: Phys.org – Physics-informed AI excels in large-scale materials discovery

As part of the Research Leaders (RL) stage in the Undergraduate Researchers initiative, we were asked to complete a research-based task. I decided to focus on a topic I’m really passionate about , The role of Artificial Intelligence and Machine Learning in building a sustainable future for agriculture. I summarized my work in a poster titled : “Addressing Problems and Shaping Sustainable Futures” where I highlighted how AI and ML technologies can help solve major agricultural challenges such as climate change, resource scarcity, and productivity improvement along with global case studies and adoption barriers. - You can check out the full text by scanning the qr code 🤳.

AI may be eating the world, but humans cannot eat AI. As billions pour into artificial intelligence, we risk neglecting the very systems that sustain life—our oceans, our food, our climate. To secure our future, we must balance investment in code with investment in carbon, building sustainable value where humanity’s needs meet technological innovation.