A SIGGRAPH 2025 report on NVIDIA's special address: how AI, scalable simulation, differentiable physics, and digital twins are launching 'Graphics 3.0' and reshaping graphics, robotics, and content creation.

SIGGRAPH 2025: Simulation, AI, and the Rise of Graphics 3.0

SIGGRAPH 2025 Special Address — AI, Simulation, and the Dawn of Graphics 3.0

AI is reshaping what’s possible in graphics. ✨

Hear NVIDIA AI research leaders Sanja Fidler, Aaron Lefohn, and Ming-Yu reveal our latest breakthroughs in computer graphics and physical AI at #SIGGRAPH2025.

SIGGRAPH 2025 Special Address

A hands-on look at Opal, Google's experimental no-code tool that turns natural language into multi-step AI workflows—research, draft and generate video without writing code.

Opal by Google — No-Code Composer for AI Mini-Apps

Introducing Opal: Google’s No-Code Tool for Composing AI Mini-Apps

Discover Opal, a new experimental tool that helps you compose prompts into dynamic, multi-step mini-apps using natural language.

Opal removes the need for code, allowing users to build and deploy shareable AI apps with powerful features and seamless integration with existing Google tools.

Try Opal → https://opal.withgoogle.com
Subscribe to Google for Developers → https://goo.gle/developers 

Speaker: Elle Zadina
Products Mentioned: Google AI

Introducing Opal

An investigative report on Hitler’s Wunderwaffen — V‑3, rocket fighters, guided bombs and giant guns — assessing technical claims, operational failures and human toll.

Wunderwaffen Exposed: Nazi Superweapons — Engineering Ambition and Human Cost

Hitler's Greatest Secrets: The Nazis' Wunderwaffen

V posledních letech Třetí říše, kdy se nad Německem stahovala porážka, vsadil Hitler na arzenál tajných zbraní, které podle něj mohly změnit průběh války. Řízené střely, létající křídla neviditelná pro radar, obří železniční děla, raketou poháněné stíhačky a projekty mezikontinentálních bombardérů byly součástí ambiciózního programu Wunderwaffen, takzvaných „zázračných zbraní“. Od hlubin tajných bunkrů až po nejmodernější hangáry spojovaly tyto projekty extrémní technologické inovace s bezprecedentní vojenskou zoufalostí.

Mezi nejúžasnějšími výtvory byl superkanón V-3 schopný ostřelovat Londýn z Francie, stroj Horten Ho 229 se stealth konstrukcí, děsivé chemické a biologické látky vyvinuté jednotkami SS a řízené střely Fritz X a X-4. Objevily se také stíhačky jako Me 163 Komet, bombardéry typu Amerika Bomber určené k útokům na New York, složená letadla Mistel naložená tunami výbušnin a oceloví giganti jako kanóny Gustav a Dora. Každý z těchto projektů, ač technicky inovativní, byl poznamenán logistickými problémy, průmyslovými omezeními a především tím, že přišly příliš pozdě, aby ovlivnily výsledek války.

Ačkoli většina z nich nikdy nedosáhla rozhodujícího bojového nasazení, mnohé z těchto technologií zanechaly stopu v poválečném letectví, vojenském inženýrství i vědě. Nacistické inženýry si po válce najaly vítězné mocnosti, jejich návrhy inspirovaly vývoj během studené války a jejich zbraně, zahalené mýty a legendami, dodnes vzbuzují fascinaci. „Největší tajemství nacistů“ neodhaluje jen rozsah Hitlerových technologických ambicí, ale také to, jak sen o nemožném vítězství dal vzniknout některým z nejpodivnějších, nejpokročilejších a nejnebezpečnějších válečných strojů v historii.

Největší tajemství nacistů: Hitlerovy zázračné zbraně

Recap of OpenAI Podcast Ep. 5 with Jakub Pachocki and Szymon Sidor: AGI progress, surprising breakthroughs, automating scientific discovery, long‑horizon reasoning, and societal implications.

AGI Progress & Automating Discovery — Insights from OpenAI Podcast Ep. 5

AGI progress, surprising breakthroughs, and the road ahead — reporting from the OpenAI Podcast Ep. 5

How close are we to automating scientific discovery? What do AI competition wins really tell us about progress toward AGI? OpenAI Chief Scientist Jakub Pachocki and researcher Szymon Sidor share inside stories—from gold medals at the International Math Olympiad to surprising leaps in reasoning—that reveal where AI is headed next.

Chapters
1:20 – From high school in Poland to AI research leaders
4:50 – Explaining AGI: technical and everyday perspectives
6:30 – Automating scientific discovery with AI
7:50 – Breakthroughs in medicine, AI safety, and alignment
10:30 – Today is a decade in the making
14:30 – Benchmark saturation and its limits
16:50 – Why math competitions matter for AI
18:15 – How models reason without tools
21:45 – Recognizing when a model can’t solve a problem
23:30 – Storytime: AtCoder competition in Japan
26:50 – How reasoning breakthroughs really happen
28:55 – What’s next for scaling and long-horizon reasoning
30:30 – What AGI will look and feel like
36:25 – Balancing trust and personal value
34:00 – Advice to high school students in 2025

AGI progress, surprising breakthroughs, and the road ahead — the OpenAI Podcast Ep. 5

Anthropic’s interpretability team opens LLM internals—visualizing circuits, finding concepts like sycophancy, landmark and numeric representations, and improving safety and trust.

Inside the AI Mind: How Anthropic Probes Model Interpretability

Interpretability: Understanding how AI models think

What's happening inside an AI model as it thinks? Why are AI models sycophantic, and why do they hallucinate? Are AI models just "glorified autocompletes", or is something more complicated going on? How do we even study these questions scientifically?

Join Anthropic's Josh Batson, Emmanuel Ameisen, and Jack Lindsey as they discuss the latest research on AI interpretability.

Read more about Anthropic's interpretability research: https://www.anthropic.com/news/tracing-thoughts-language-model 

Sections:
Introduction [00:00]
The biology of AI models [01:37]
Scientific methods to open the black box [6:43]
Some surprising features inside Claude's mind [10:35]
Can we trust what a model claims it's thinking? [20:39]
Why do AI models hallucinate? [25:17]
AI models planning ahead [34:15]
Why interpretability matters [38:30]
The future of interpretability [53:35]

A frontline dispatch of the Ardennes Offensive (Battle of the Bulge): the surprise December 1944 German strike, the human stories, strategic choices and why the gamble failed.

Ardennes Offensive 1944: Hitler’s Last Gamble Unfolded

Battle of the Ardennes: The Last Gamble of Nazi Germany

V prosinci 1944 se klidné a zasněžené Ardeny staly dějištěm poslední velké německé ofenzívy na západní frontě. Pod krycím názvem Wacht am Rhein Hitler tajně shromáždil více než 200 000 vojáků a téměř 1 000 tanků, aby překvapivým útokem zasáhl spojenecké síly, prolomil jejich linie, dobyl Antverpy a změnil průběh války. Útok, zahájený za úsvitu 16. prosince, zastihl Američany nepřipravené a způsobil počáteční kolaps fronty, kdy byly vesnice srovnány se zemí, celé pluky obklíčeny a neúprosná zima se stala dalším nepřítelem.

Ofenzíva byla poznamenána epizodami brutality a extrémního odporu. SS Kampfgruppe Peiper postupovala s velkou silou a stála za masakrem v Malmedách, zatímco v Bastogne 101. výsadková divize USA odolávala prudkému obléhání a slavně odmítla kapitulaci slovy „Nuts!“. Boje se rozšířily do lesů, vesnic a zledovatělých silnic, s tankovými souboji mezi Panthery a Tiger II proti lehčím Shermanům a s diverzními operacemi komand v přestrojení, jejichž cílem bylo zasít chaos. Chlad, mlha a nedostatek paliva zpomalily německý postup, zatímco morálka spojenců se s každým dnem bojů zpevňovala.

Když se 23. prosince vyjasnilo, spojenecké letectvo bez slitování útočilo na německé kolony, přerušovalo zásobovací linie a blokovalo ústupové cesty. Příchod Pattonovy 3. armády prolomil obklíčení Bastogne a spojenecké protiútoky obklíčily německé jednotky, které začaly chaoticky ustupovat směrem k Rýnu. Bitva v Ardenách si vyžádala více než 180 000 obětí a znamenala konec útočné schopnosti Třetí říše na Západě, čímž zpečetila cestu k definitivní porážce Německa v roce 1945.

Bitva v Ardenách: Poslední pokus nacistického Německa

How design-driven brands use visual systems, employee advocacy, and creative ops to boost recall, conversions, and brand affinity—practical tactics and playbooks.

Why Visual-First Brands Win: Design as a Growth Engine

Visual first: How design-driven brands are outperforming competitors

Join industry-experts from Canva, Exos, and Ricoh as you learn more about: 

- Visual communication trends across markets
- Turning every employee into a brand storyteller
- Data: The impact of strong visual brands on business growth
- Building an agile design infrastructure

Discover how NVIDIA RTX PRO Blackwell GPUs revolutionize video editing with real-time 4K multi-cam playback, precise color grading, and AI-powered tools to elevate your creative workflow.

Unlock Faster, Smarter Video Editing with NVIDIA RTX PRO Blackwell GPUs

Video Editing with NVIDIA RTX PRO: Elevate Your Workflow with Power and Precision

Edit faster, color deeper, and create smarter with NVIDIA RTX PRO Blackwell GPUs. 🚀 From native 4:2:2 10-bit support to AI-powered effects and lightning-fast encoding, discover how RTX PRO transforms multi-cam, high-bit-depth, and AI video editing workflows.

Learn more: https://www.nvidia.com/en-us/products/workstations/

Video Editing with NVIDIA RTX PRO

Explore NVIDIA Research's groundbreaking insights from SIGGRAPH 2025 on the convergence of AI, graphics, and robotics, highlighting GPU-driven advances and the future of 3D intelligence.

NVIDIA Research at SIGGRAPH 2025: Pioneering AI, Graphics, and Robotics Innovation

Highlights From NVIDIA Research's Special Address at SIGGRAPH 2025: The Future of Graphics, AI, and Robotics

At SIGGRAPH 2025, the premier computer graphics conference, NVIDIA Research leaders delivered a special address highlighting the graphics and simulation innovations enabling physical and spatial AI. 

Learn more: https://blogs.nvidia.com/blog/physical-ai-research-siggraph-2025/

Highlights From NVIDIA Research's Special Address at SIGGRAPH 2025

Discover how NVIDIA's neural reconstruction and Gaussian splatting techniques are creating photorealistic 3D simulations that transform robotics and autonomous vehicle development.

Revolutionizing Robotics and Autonomous Vehicles with Neural Reconstruction

Neural Reconstruction for Robotics and Autonomous Vehicles: The Future of Physical AI Simulation

Experience the future of robotic and autonomous vehicle development with the latest in Gaussian splatting and generative AI techniques. High-fidelity sensor data is critical to developing safe autonomous systems, however, accurately rendering the diversity of the real world in simulation is a tremendous challenge. This video demonstrates how NVIDIA Omniverse NuRec neural reconstruction libraries bring the real world into simulation, using multi-sensor data to achieve photorealistic environments essential for testing and validating robotics systems.

Read the tech blog: https://developer.nvidia.com/blog/how-to-instantly-render-real-world-scenes-in-interactive-simulation/

#NVIDIAOmniverse
#OmniverseNuRec
#GaussianSplatting
#SIGGRAPH2025

Neural Reconstruction for Robotics and Autonomous Vehicles

Discover how NVIDIA Cosmos bridges the sim-to-real gap by generating realistic training data with AI and computer graphics, speeding up robot learning for real-world applications.

How NVIDIA Cosmos Accelerates Robot Training with AI-Driven Simulation

Bridging the Sim-to-Real Gap for Accelerated Robot Training

Training robots to navigate the complexities of the real world requires an immense amount of data, a challenge that has traditionally been slow and expensive. Advances in the technology powering NVIDIA Cosmos enable developers to accelerate workflows for generating realistic training data at the scale required for rapid development. This breakthrough helps robots learn faster and safer, leading to quicker real-world testing, iteration, and innovation. 

Discover how Cosmos is closing the simulation-to-reality gap and driving the future of robotics forward.

Read the tech blog: https://developer.nvidia.com/blog/maximize-robotics-performance-by-post-training-nvidia-cosmos-reason


#NVIDIACosmos
#SIGGRAPH2025

Discover how NVIDIA's generative AI breakthroughs are transforming digital twin creation for physical AI, enabling faster, more accurate simulations to train autonomous systems safely and effectively.

Revolutionizing Physical AI with AI-Powered Material Generation

Research Advances in AI-Assisted Material Generation for Physical AI

Digital twins are vital for the training and testing of physical AI that can understand and perform complex actions in the real world. But building digital twins is slow and tedious. NVIDIA’s latest generative AI and rendering research showcases the potential to dramatically accelerate and scale these workflows. 

These breakthroughs combine NVIDIA Cosmos World Foundation Models and advanced rendering technologies to allow developers to pair AI assistants with experts at the heart of digital twin and physical AI development. In these future workflows, experts will engage with AI assistants using simple language to quickly generate realistic, physically accurate base materials for 3D models. They can then fine-tune generated materials, adjusting properties such as roughness and texture to better reflect the real world.

This research, along with the workflows it enables, showcases the potential to quickly transform 3D models into large-scale, simulation-ready digital twins that can be used to train safe and efficient physical AI.

Read the research paper: https://s2025.conference-schedule.org/presentation/?id=papers_1278&sess=sess123

#OpenUSD
#DigitalTwin
#NVIDIAOmniverse
#IndustrialAI
#SIGGRAPH2025

AI and Technology News

Latest

AI News