Weather forecasting usually means huge computers grinding through equations that describe the atmosphere. Anima Anandkumar helped build a very different route. The IIT Madras graduate and Caltech professor has now been named to TIME’s 2026 list of the 100 most influential people in artificial intelligence, with TIME highlighting her work on AI systems that model physical processes at extraordinary speed.
Her best-known weather project, FourCastNet, showed that an AI model could produce high-resolution forecasts far faster than conventional numerical forecasting. The story has become even more timely in August 2026. Anandkumar has launched a physics-focused AI company, joined a United Nations scientific advisory body this year, and is pushing neural operators beyond weather into chips, fusion and scientific simulation. TIME published its 2026 profile of Anima Anandkumar on August 27.
Who Is Anima Anandkumar And Why Is TIME Recognising Her?
Anandkumar graduated from IIT Madras with a BTech in Electrical Engineering in 2004 before earning a PhD in Electrical and Computer Engineering from Cornell University. She later worked at MIT, the University of California Irvine, Amazon Web Services and NVIDIA. She is now Bren Professor of Computing and Mathematical Sciences at Caltech. IIT Madras also gave her its Distinguished Alumnus Award in 2024, as recorded in her IIT Madras alumni profile.
TIME’s 2026 profile focuses on a thread that runs through much of her research: using AI to model physics without repeating every expensive calculation from scratch. TIME says some of her techniques have made physical simulations more than a million times faster than earlier methods, while weather applications have run thousands of times faster than traditional systems.
TIME also announced the 2026 TIME100 AI list through its official X account, calling it a collection of the world’s most influential people in artificial intelligence. The official TIME X announcement can be embedded here. The post went live on August 27 alongside the new TIME100 AI cover announcement.
How Does FourCastNet Predict Weather So Much Faster?
Traditional numerical weather prediction starts with the current state of the atmosphere and repeatedly solves complex physics equations across a global grid. That work uses hefty computing resources. FourCastNet takes another path. It learns patterns in atmospheric data and predicts how weather variables evolve from one time step to the next.
What Neural Operators Actually Do
Anandkumar helped pioneer neural operators, models designed to learn relationships between continuous physical fields. FourCastNet used an Adaptive Fourier Neural Operator, or AFNO, to work with global weather data at 0.25-degree resolution. NVIDIA’s FourCastNet documentation says the system forecasts variables including wind speed, precipitation, and atmospheric water vapour.
The speed changes what forecasters can try:
- A forecast can be generated in seconds rather than waiting hours for a large conventional run.
- Researchers can run many slightly different scenarios and compare how storms may develop.
- Large ensembles can improve estimates of uncertainty around hurricanes, heatwaves and other extreme events.
- Faster inference also lowers the computing barrier for experiments that once demanded major supercomputing resources.
Caltech’s research group has described its AI weather work as roughly 45,000 times faster while retaining comparable forecast accuracy. TIME earlier reported that FourCastNet could produce a week-long forecast in under two seconds. NVIDIA’s newer FourCastNet 3 model goes further, producing a 60-day global forecast at 0.25-degree resolution in under four minutes on one GPU.
Why Her Work Is Bigger Than Weather Forecasting
The recognition arrives during a busy period for Anandkumar. In March 2026, she joined the UN Secretary-General’s Scientific Advisory Board, a 15-member body that advises UN leadership on developments across science and technology. Caltech detailed the appointment in its official announcement.
Then came a new company. Reuters reported this week that Anandkumar and Benedikt Jenik launched Accelerated Understanding Inc., which is building AI around physics rather than language. The company says one of its models handled five trillion pieces of data in a single prompt. Proposed uses include semiconductor design, robotics, extreme-weather prediction and energy exploration.
That gives Anandkumar’s weather work a wider context. FourCastNet was not simply an attempt to make tomorrow’s forecast arrive sooner. It became evidence that AI could learn useful physical behaviour, then perform some simulations at speeds that alter how researchers test possibilities. Her neural-operator research has also been applied to nuclear-fusion plasma, medical-device design and components used in computing.
There is an important limit, though. Faster AI does not remove the need for observations, meteorologists or physical validation. Forecast quality still depends on good input data, testing across unusual events and careful comparison with established systems. The useful gain is speed plus scale, especially when thousands of possible outcomes need to be explored quickly.
FAQs About Anima Anandkumar And FourCastNet
Who is Anima Anandkumar?
She is an IIT Madras graduate and Caltech professor known for AI-driven scientific modelling research.
Why is Anima Anandkumar in TIME100 AI 2026?
TIME recognised her influential work using AI to accelerate weather forecasting and other scientific simulations.
What is FourCastNet?
FourCastNet is an AI weather model designed to produce high-resolution global forecasts using neural-operator technology.
How fast is FourCastNet compared with traditional forecasting?
Caltech says AI weather models can run tens of thousands times faster than conventional forecasting.
What is Anima Anandkumar working on now?
She is advancing physics-focused AI through Caltech, UN advisory work and her launched technology company.
