(Shutter stock AI photo)
AI is in the forefront of the weather forecast as the next big jump, and rightly. Recent development promises fast and more accurate predictions of severe weather events. In August 2024, NVIDIA’s Storsts Model Draws Heads Looks for unprecedented speed and ability to predict precision storms.
Right now, Google Deep Mind Graphicist Model Draws Attention to Improve Traditional Systems In the selected weather prediction benchmark. Such AI -based models have given rise to hope that AI communities can help better prepare for floods, hurricanes and other climate disasters.
However, when the devastating floods hit Texas in July 2025, the promise was largely unrealistic. As the flood waters increased, AI models deprived of important indications that could help prevent damage.
The Trump administration may reduce the $ 2.2 billion deduction deductions by the Trump administration to support AI research and infrastructure needed to help models such as Storcast and Graphics reach their full potential.
Texas floods painfully cleansed the boundaries of today’s AI models. Since the storm intensified during the end of July 4, many modern systems that failed to predict the intensity of something. It was a traditional high resolution model, not AI-driven, who gave the most accurate picture of danger.
“All those new Fancy AI model? They also lost it,” said Daniel Swin, a climate scientist at the California Institute for Water Resources. Live YouTube Talk July 7.
(Best background/shutter stock)
Swin pointed out that while the AI systems have grown rapidly, when they talk about predicting extreme weather, they develop rapidly in small geographical areas. Heavy rainfall was extremely local in the mountainous country in Texas, and most AI models were not designed to handle such details.
These were the NOAA’s high resolution tools for predictions, which were specifically designed to imitate the storm behavior, which performed well and gave predictors to the predictions of serious problems.
Swin also warned that these models could not be too long. The proposed budget will close many research centers responsible for producing these highly performing models. These decisions can eliminate some reliable forecasting tools in the same way that climate disasters are becoming more often. In his view, the risk is not only losing the earth, but the eraser systems that already work.
Noaa has offered a more confident diagnosis, for its part. In a statement, spokesman Kim Doster said that despite the proposed budget cuts, the agency’s research and prediction priorities will remain. He already released modern modernization efforts, AI played a central role in the future prediction.
Commerce Secretary Howard Lotank, who oversees Noaa, echoed the message that the administration is committed to providing sharp, sharp weather data through modern technologies. Doester added that NoaA is working closely with scientists to improve storm mapping, reduce warning hours and provide better data to the public.
The agency’s confidence is sitting seriously, which is at stake. Predictors’ tools who performed well during the Texas floods came from NoaA’s research programs. Is Programs may be influenced by deductions in the proposed budget. Although the AI has become a major priority, it is unclear how far it can be done if the basic systems and teams behind it lose support.
(Deva Heritawan/Shutter Stock)
It is worth noting that NOAA is developing its AI tools, including Project EagleDesigned to improve the forecast of high -impact events, but the system was allegedly not worked, and could not help in response during the Texas flood.
Noaa was not alone in the failures of Texas. Although the agency has highlighted a lot of light, many of today’s advanced AI predictions are being prepared in the private sector. Companies are running to develop a global prediction model using deep learning and wide datases. But despite their technical edge, the system also lost the mark.
“Local rain forecast is very difficult, and most of the now used AI models have not been the focus,” said Russia Schumakar, a state meteorological expert in Colorado. This limit became clear when many private sector models also failed to detect the scale and time of the flood.
One of the problems is that most AI’s prediction systems learn from historical weather data. But events like the 2025 Texas floods are rare and extreme, which means that there are not many past examples to learn. Without this point point, it is difficult to spot the models when there is really something out of the ordinary.
Another challenge is a resolution. Many AI models are designed to capture wider global patterns, not such local storms that cause floods. This limits their ability to detect rapid motion and high influence in certain areas. Unless AI tools can work on a better scale and add real time data more effectively, their predictions will decrease.
(Quality Stock Arts/Shutter Stock)
Nevertheless, some experts believe that AI has an important role in the future of prediction. Former Acting Noaa Administrator, Tim Gallowet recently argued that national seasonal services should increase its use of AI in environmental, maritime and hydroological models to improve accuracy.
However, Texas floods served as a reality examination. He showed that even the latest models, whether it be government labs or private companies, still encountered a fast moving, local disaster. AI is promising, but it is not ready to stand on its own.
For now, the space is clear. Improving forecasts will take more than better technology. It will depend on the system, data, and a stable investment in people who help prepare communities before the next crisis.
Relevant
