Importance of Physical Understanding in AI Development

🚨 Elon Musk and Nvidia CEO Jensen Huang are urging students to look beyond just learning how to code. As AI gets better at handling repetitive tasks, both believe the real advantage will come from understanding how the world works. Through physics and math. Jensen Huang recently said that if he were graduating today, he'd focus on physics. He explained that future AI systems will need to work with the physical world, not just digital spaces. This means knowing how things move, how forces interact, and how systems behave in real life. Elon Musk has echoed the same idea. When asked about useful skills for the future, he pointed to physics, backed by math. At Tesla and SpaceX, his thinking is rooted in solving problems from the ground up using core principles, not just following existing methods. They’re not saying coding is useless. It still matters. But the next big opportunities will go to people who understand the systems AI is meant to model, control, and improve. In simple terms, learn how the world really works. Study the tough stuff. Physics and math build the kind of thinking that machines can’t easily replace. ------- Do you agree?

A ball dropping on wooden blocks might be the future of AI. Sounds simple. It isn’t. When researchers simulate a ball falling onto stacked woodblocks, they’re not playing with physics toys — they’re teaching AI cause and effect. And the data backs it up : • Physics-informed AI models can predict object interactions 10–100× faster than traditional physics engines once trained • Deep learning models trained on physical simulations have shown 30–50% error reduction in collision and stability prediction compared to rule-based systems • Reinforcement-learning agents trained in simulated environments transfer to real-world robotics with up to 80–90% task success after fine-tuning • Studies from DeepMind, OpenAI, and MIT show AI systems with learned “intuitive physics” outperform symbolic models on future-state prediction tasks What’s really happening: • The AI learns what will fall before it falls • It predicts chain reactions instead of reacting after the fact • It develops something close to human physical intuition The same intuition that lets a child know: “That tower will collapse if I touch it.” Why this matters: 🏗️ Engineering – Predict structural failure before construction 🤖 Robotics – Robots that understand balance, force, and contact 🚗 Autonomy – Millisecond-level prediction of collisions and debris 🎮 Simulation & games – Physically believable, adaptive worlds 🧠 AI research – A step toward real-world common sense The real breakthrough isn’t better graphics or faster math. It’s this shift: From calculating physics To understanding physics Once AI understands how the world behaves… everything it touches becomes smarter. The future of AI might not start with language. It might start with a falling ball. Do you think physical intuition is the missing layer for general AI? #ArtificialIntelligence #Simulation #Robotics #Engineering #FutureOfAI #MachineLearning #TechTrends #DeepLearning

Most robotics companies don’t fail because their hardware is weak. They fail because their AI does not survive reality. The moment conditions drift — lighting, clutter, wear, timing, human interaction — perception breaks, reasoning degrades, and recovery logic explodes in complexity. What looks like an AI problem quickly becomes a business problem: stalled deployments, high integration cost, and products that never scale. I’ve been writing a growing document on this exact gap: How to Build Physical AI https://lnkd.in/gZXk9TJX This is an AI-first blueprint, written for teams building real robots, not demos. The premise is simple: Physical AI must reason over time, geometry, and uncertainty — not just predict from pixels. In the document, we outline how modern AI stacks fall short and what replaces them: • World models grounded in physics, not frame-by-frame perception, so policies generalize across sites and conditions • Long-horizon reasoning that remains stable as the environment unfolds in real time • Edge-native AI architectures that close the loop between perception, decision, and action without cloud dependence • Learning systems that compound, instead of resetting every time you introduce a new SKU, layout, or task variant For CEOs, this directly impacts rollout velocity, gross margins, and customer confidence. For CTOs, it determines whether your AI roadmap compounds or collapses under edge cases. Physical AI is not about adding more models. It’s about building an intelligence layer that understands the physical world well enough to act in it — reliably, repeatedly, and at scale. This document will continue evolving as the field matures. If you’re building AI that has to run on real machines, in real environments, I’d love your perspective. #AI #PhysicalAI #Robotics #Autonomy #EngineeringLeadership Dragomir Anguelov Aditya Jami Steve Cousins Oliver Cameron

Jensen Huang and Elon Musk both say the same thing: That you should skip coding. Study physics and math instead. Here’s why: AI can: - Generate perfect functions. - Optimize algorithms. - Debug code. And it does it better than many developers. But AI lacks understanding of how the physical world actually works. It can’t: - Anticipate how complex systems behave in unpredictable real-world conditions. - Infer cause-and-effect from incomplete or noisy physical data. - Integrate multiple layers of physical constraints into a coherent solution. Meanwhile, physical AI is accelerating: - Drones adjust for wind and turbulence to stay on course. - Warehouse bots calculate load balance before lifting. - Surgical robots adapt to tissue resistance in real time. This creates a massive opportunity gap. While developers panic about AI replacing programmers and code, the real question that matters is different: Who can model reality? Like Tesla and NVIDIA, the biggest companies of the future will be the ones who understand: - How systems actually behave - Why things break under real conditions - How to reason from first principles Physics understanding is the competitive moat for building solutions that actually take the world into a new era. The physics and science first thinkers will build the companies owning the next decade where AI meets: - Aerospace & Space Systems - Autonomous & Robotics Systems - Advanced Manufacturing & Industrial Automation - Renewable Energy & Energy Storage - Transportation & Mobility Systems - Climate Tech & Environmental Systems - Medical Devices & Biomechanics - Defense & Aerospace Simulation The future belongs to people who can model the physical world.

Jensen Huang called physical AI a $50 trillion opportunity. He is not wrong. AI is moving beyond software agents into physical systems. Robotics, multimodal models, and edge computing are converging. The frontier is not cognition, it’s coordination. This is not about humanoid robots in your living room. This is about AI that can see, touch, and manipulate the physical world at an industrial scale. The software AI wave trained models on text and images. Physical AI trains on sensor data, motor controls, and real-world physics. Every factory, warehouse, and construction site becomes a training ground. The companies winning this will not be the ones building the smartest chatbots. They will be the ones teaching AI to operate forklifts, assemble products, and navigate complex environments. Physical AI solves the constraint that has limited AI impact, the digital-physical gap. Software AI can analyze your supply chain. Physical AI can fix it. Watch for AI companies partnering with hardware manufacturers. Edge computing startups targeting industrial use cases. Robotics companies hiring former OpenAI engineers. The next $50 trillion will be built by AI that can touch the real world.

Deep stuff! We uncovered a startling link between #entropy, a bedrock concept in #physics, and how #AI can discover new ideas without stagnating. In an era where reasoning models can reflect on problems for days at a time (rather than generating quick, single-step solutions), our study shows how semantic entropy (the spread of meanings) and structural entropy (how evenly its links between concepts generated by the AI are distributed) together hold the secret to ongoing exploration as the model thinks through a problem. Specifically, we measured structural entropy using Von Neumann graph entropy (applied to the adjacency Laplacian), while semantic entropy came from a similarity-based embedding deep language embedding matrix. The key insight? Although semantic entropy consistently outpaces structural entropy, they remain in a near-critical balance—fueling "surprising edges" that introduce relationships between distant concepts. This mirrors physical systems on the brink of a phase transition, where a little bit of "disorder" keeps the process dynamic yet avoids chaos. The result is an AI that doesn’t just keep pace with known solutions but actively creates new pathways of thought over extended “thinking” sessions. As reasoning models become ever more capable—undertaking extended, multi-day "thought processes"—understanding fundamental principles is crucial. By weaving these insights into reinforcement learning strategies, we can reward models not just for correctness, but for venturing into novel conceptual ground. This opens the door to AI systems that actively cultivate new insights, rather than settling into narrow patterns or endlessly rehashing the same knowledge. Going Deeper When physicists describe entropy, they refer to the measure of "disorder" in a system: the number of ways particles can rearrange without altering the system’s energy. Yet entropy transcends molecules and heat. In this research, it emerges as the engine that drives AI reasoning models to keep generating fresh ideas over extended periods. The observed dynamics as the AI thinks about a problem reflects self-organized criticality—a state where systems hover between rigid order and random chaos. Much like a sand pile teetering on the edge of collapse, the AI preserves enough organizational structure to remain coherent, yet stays flexible enough to generate unexpected leaps in meaning. The fraction of "surprising edges" remains stable, offering evidence that the model naturally integrates new, distant ideas without toppling into confusion.

I recently spoke to Gartner about what is next in #AI. Here are my thoughts: We have seen impressive progress in #llm by scaling data and compute. Will this continue to hold? Yes, I believe so, but most of those gains will be in reasoning tasks where we have precise metrics to measure uplift, as well as the ability to have synthetic data to train further, and also the freedom to trade off computation for accuracy at test time. This is seen in the recent o1 model. For reasoning tasks, we will also be able to remove hallucination when we can construct accurate verifiers that can certify every statement that #llm makes. We have been doing this in our Leandojo project for mathematical theorem proving. However, there is one area of reasoning where #llm will never be good enough: understanding the physical world. This is because language is only high-level knowledge, and cannot simulate the complex physical phenomena needed in many applications. For instance, LLMs can talk about playing tennis or look up a weather app, but they cannot internally simulate any of these processes. While images and videos can help improve their knowledge of the physical world, models like Sora learn physics by accident, and hence, still produce physically wrong outputs. How can we overcome this? By teaching AI physics from the ground up. We are building AI models that are trained in a physics-informed manner at multiple scales. They are several orders of magnitude faster than traditional simulations, and can also generate novel designs that are physically valid. You can watch some of those examples in my recent TED talk.

From CES this week, one thing is clear: we are moving into the era of physical AI — intelligence that operates in the real world.    Robotics, including humanoid and non-humanoid systems, are getting a lot of attention right now. This is familiar territory for Autodesk. We have decades of experience working with manufacturing, AI, and industrial design leaders who build in the physical world.    MarketWatch recently explored this momentum and included some of my perspective: https://lnkd.in/e_DN9HwC    Progress will not come from machines that just look like us, nor just language. It will come from AI that understands physics, objects, and three-dimensional space. That’s why work on world models, like what Fei-Fei Li and others are doing, matters. These systems learn from sensory data to build a usable understanding of their environment.    Physical AI will change how every industry that makes things designs, simulates, and executes. That is core to Autodesk’s mission, and I am optimistic about what is ahead.    Who is ready to put physical AI to work across everything we design and build? 

You cannot train AI on reality alone anymore. There is not enough of it. Jensen Huang explains why NVIDIA built Cosmos, an AI world model that generates synthetic training data grounded in physics. The problem is simple. Teaching physical AI like robotics requires vast amounts of diverse interaction data. Videos exist, but not nearly enough to capture the variety of situations robots will encounter. So NVIDIA transformed compute into data. Using synthetic data generation grounded by laws of physics, they can selectively generate training scenarios that would be impossible to capture otherwise. The example Huang shows is remarkable. A basic traffic simulator output gets fed into Cosmos. What emerges is physically plausible surround video that AI can learn from. This solves a fundamental limitation. You cannot train autonomous systems on every possible scenario by recording reality. There are not enough cameras or time. But you can simulate physics accurately enough that AI trained on synthetic data generalises to real environments. This applies beyond robotics. Any AI learning physical interactions, from manufacturing to logistics to infrastructure monitoring, faces the same data scarcity problem. Synthetic data generation grounded in physics laws is how you create training sets reality cannot provide. The organisations building AI for physical systems will either master synthetic data generation or get limited by whatever reality they can record. Watch the full presentation to hear Huang explain how Cosmos generates training data for physical AI. What physical AI application needs synthetic data because reality cannot provide enough examples? #AI #SyntheticData #Robotics #NVIDIA #MachineLearning

🚀 The Physics AI market is no longer just a concept… it’s a reality. "How do you simulate complex physical systems when traditional methods are too slow or too expensive?" In recent months, we’ve seen a wave of startups and initiatives focused on combining physics and artificial intelligence. And the trend is clear: demand for solutions in this space is rapidly increasing. From automotive, aerospace, and energy sectors to oil & gas, civil engineering, and embedded systems, Physics-Informed AI has the potential to revolutionize how we model, simulate, and predict physical behavior in complex systems. But make no mistake, building solutions in this field is far from simple. It takes more than machine learning or software engineering expertise. A deep understanding of the physical and mathematical foundations behind the problems is essential. About me: I specialize in design patterns for Physics-Informed Neural Networks (PINNs) and have worked on solutions that include: - Modeling and training of PINNs and Physics-Informed Extreme Learning Machines; - Application of the Theory of Functional Connections to improve model performance and stability; - Development of Surrogate Models to accelerate complex simulations; Integration of simulation data (CFD, FEM, etc.) with AI models; - Use of tools like NVIDIA Omniverse and Modulus/PhysicsNeMo for interactive physics pipelines; - Generation of physics-based datasets by extracting physical features from scientific or industrial videos 🎯 For technical leaders building the next generation of physics-driven systems, I bring hands-on experience where physics meets machine learning, helping bridge complex simulation workflows with AI. In recent projects, my contributions have included: - Architecting and implementing custom solutions informed by physical laws - Leading applied research initiatives focused on performance and stability - Advising internal teams on strategy, infrastructure, and technical decision-making - Supporting integration of physics-based models, simulators, and data frameworks with AI pipelines If your team is looking to operationalize Physics AI, whether accelerating development, reducing simulation costs, or designing a scalable pipeline, I’d be glad to share insights and explore how we can move further, faster. Let’s connect and build something that truly respects both the science and the system. #physicsai #pinns #machinelearning #engineering #physics #simulation #ai #consulting #deeptech #computationalmodeling #nvidia #tfc #surrogatemodels