AI has started to emerge as one of the most effective technologies being used in cosmology lately. The power of machine learning technologies is seen when analysing galaxies and models that predict the evolution of the universe. Yet a recent study has uncovered an unexpected complication. When researchers trained an AI model on the standard cosmological framework that explains the universe, the system became exceptionally efficient at recognising familiar patterns, but significantly less effective at detecting genuinely new physics. The finding highlights a growing challenge in scientific AI: the very knowledge that makes these systems powerful may also make them resistant to revolutionary discoveries. As astronomers prepare for an unprecedented flood of data from next-generation observatories, understanding this limitation could prove as important as the technology itself.
How AI learned the rules governing the universe
Researchers from the Department of Astrophysical Sciences, Princeton University, state that modern cosmology is built around the Lambda Cold Dark Matter (ΛCDM) model. This framework explains the large-scale structure and evolution of the universe. It successfully describes phenomena ranging from galaxy formation to cosmic expansion and remains the prevailing model in cosmological research.To accelerate scientific investigations, researchers trained neural networks using simulations generated under ΛCDM assumptions. The approach relied on a machine-learning technique known as transfer learning, in which an AI first learns broad patterns from simpler datasets before being adapted to tackle more specialised tasks. including massive neutrinos, modified gravity, and primordial non-Gaussianities, can enable inference with significantly fewer beyond-ΛCDM simulations.However, we also show that negative transfer can occur when strong physical degeneracies exist between ΛCDM and beyond-ΛCDM parameters. We consider various transfer architectures, finding that including bottleneck structures provides the best performance. Our findings illustrate the opportunities and pitfalls of foundation-model approaches in physics: pretraining can accelerate inference, but may also hinder learning new physics.The results were impressive. Scientists found that transfer learning dramatically reduced the number of computationally expensive simulations required to analyse alternative cosmological models. In some cases, the approach lowered computing demands by more than an order of magnitude, potentially saving years of processing time and significant research costs.The study’s lead researchers demonstrated that AI systems can rapidly identify subtle relationships within vast cosmological datasets, making them invaluable for future projects that will generate petabytes of observational information.
The unexpected problem physicists did not anticipate
The same prior knowledge that made the AI efficient also created a significant weakness. Researchers discovered that when the neural network became highly familiar with ΛCDM-based patterns, it sometimes struggled to recognise signals that deviated from those expectations. In essence, the system developed a form of scientific bias. It was inclined towards interpreting new information based on what it had learned rather than being receptive to possibilities of change and innovation.This poses a major problem for cosmologists in the search for proof of anything that does not conform to the standard model such as modified gravity, changing dark energy, and the effects of massive neutrinos.According to the researchers, transfer learning can become so effective at recognising familiar structures that it inadvertently suppresses the very anomalies scientists hope to discover. The challenge mirrors a long-standing issue in human science. Scientists may start their data analysis with biases based on existing theories. According to the research, AI models might be subjected to the same bias if they are trained on existing paradigms.
Why the findings may shape the future of cosmology and artificial intelligence
The discovery is important for the future of astronomy, since new telescopes and surveys will produce enormous amounts of data in the field. Nevertheless, the study highlights that future AI models should be trained to remain open-minded.Therefore, scientists might be required to create approaches that make AI sensitive to anomalies instead of being biased towards existing theories.This challenge extends beyond cosmology. Across physics, scientists are increasingly exploring AI not merely as a data-processing tool but as a mechanism for uncovering entirely new laws of nature. Recent research has shown that machine-learning systems can identify previously hidden physical relationships in complex plasma systems while maintaining interpretability, demonstrating AI’s potential as a genuine discovery engine.As physicist Justin Burton an Emory professor of experimental physics and senior co-author of the paper told The Mirage regarding AI-driven discoveries in plasma physics:”We showed that we can use AI to discover new physics,” says Justin Burton, “Our AI method is not a black box: we understand how and why it works. The framework it provides is also universal. It could potentially be applied to other many-body systems to open new routes to discovery,” he and his co authors noted in the study ‘Physics-tailored machine learning reveals unexpected physics in dusty plasmas’The new cosmology study adds an important caveat to that optimism. AI can accelerate scientific discovery, but only if researchers ensure that the systems remain capable of questioning the assumptions they were trained to understand.