AI may not need massive training data after all

Science News from research organizations AI may not need massive training data after all Building AI more like the human brain may matter far more than flooding it with data. Date: January 4, 2026 Source: Johns Hopkins University Summary: New research shows that AI doesn’t need endless training...

Science News from research organizations AI may not need massive training data after all Building AI more like the human brain may matter far more than flooding it with data. Date: January 4, 2026 Source: Johns Hopkins University Summary: New research shows that AI doesn’t need endless training data to start acting more like a human brain. When researchers redesigned AI systems to better resemble biological brains, some models produced brain-like activity without any training at all. This challenges today’s data-hungry approach to AI development. The work suggests smarter design could dramatically speed up learning while slashing costs and energy use. Share: Facebook Twitter Pinterest LinkedIN Email FULL STORY Researchers found that AI built with brain-inspired designs can mimic human brain activity before seeing any training data. The discovery challenges the idea that bigger datasets and more computing power are the only paths to smarter AI. Credit: Shutterstock New research from Johns Hopkins University shows that artificial intelligence systems built with designs inspired by biology can begin to resemble human brain activity even before they are trained on any data. The study suggests that how AI is structured may be just as important as how much data it processes. The findings, published in Nature Machine Intelligence, challenge the dominant strategy in AI development. Instead of relying on months of training, enormous datasets, and vast computing power, the research highlights the value of starting with a brain-like architectural foundation. Rethinking the Data Heavy Approach to AI "The way that the AI field is moving right now is to throw a bunch of data at the models and build compute resources the size of small cities. That requires spending hundreds of billions of dollars. Meanwhile, humans learn to see using very little data," said lead author Mick Bonner, assistant professor of cognitive science at Johns Hopkins University. "Evolution may have converged on this design for a good reason. Our work suggests that architectural designs that are more brain-like put the AI systems in a very advantageous starting point." Bonner and his colleagues aimed to test whether architecture alone could give AI systems a more human-like starting point, without relying on large-scale training. Comparing Popular AI Architectures The research team focused on three major types of neural network designs commonly used in modern AI systems: transformers, fully connected networks, and convolutional neural networks. They repeatedly adjusted these designs to create dozens of different artificial neural networks. None of the models were trained beforehand. The researchers then showed the untrained systems images of objects, people, and animals and compared their internal activity to brain responses from humans and non-human primates viewing the same images. Why Convolutional Networks Stood Out Increasing the number of artificial neurons in transformers and fully connected networks produced little meaningful change. However, similar adjustments to convolutional neural networks led to activity patterns that more closely matched those seen in the human brain. According to the researchers, these untrained convolutional models performed on par with traditional AI systems that typically require exposure to millions or even billions of images. The results suggest that architecture plays a larger role in shaping brain-like behavior than previously believed. A Faster Path to Smarter AI "If training on massive data is really the crucial factor, then there should be no way of getting to brain-like AI systems through architectural modifications alone," Bonner said. "This means that by starting with the right blueprint, and perhaps incorporating other insights from biology, we may be able to dramatically accelerate learning in AI systems." The team is now exploring simple learning methods inspired by biology that could lead to a new generation of deep learning frameworks, potentially making AI systems faster, more efficient, and less dependent on massive datasets. RELATED TOPICS Mind & Brain Intelligence Social Psychology Consumer Behavior Psychology Computers & Math Computers and Internet Computer Modeling Neural Interfaces Artificial Intelligence RELATED TERMS Workaholic Artificial intelligence Robot Mathematical model Quantum computer Brain Human brain Psychology Story Source: Materials provided by Johns Hopkins University. Note: Content may be edited for style and length. Journal Reference: Atlas Kazemian, Eric Elmoznino, Michael F. Bonner. Convolutional architectures are cortex-aligned de novo. Nature Machine Intelligence, 2025; 7 (11): 1834 DOI: 10.1038/s42256-025-01142-3 Cite This Page: MLA APA Chicago Johns Hopkins University. "AI may not need massive training data after all." ScienceDaily. ScienceDaily, 4 January 2026. . Johns Hopkins University. (2026, January 4). AI may not need massive training data after all. ScienceDaily. Retrieved July 13, 2026 from www.sciencedaily.com/releases/2025/12/251228074457.htm Johns Hopkins University. "AI may not need massive training data after all." ScienceDaily. www.sciencedaily.com/releases/2025/12/251228074457.htm (accessed July 13, 2026). Explore More from ScienceDaily RELATED STORIES AI Could Uncover New Physics Faster but There’s a Surprising Catch June 11, 2026 — Scientists found that transfer learning can make the search for new physics in the universe much faster, slashing the need for expensive simulations. Yet the approach can backfire when AI relies too ... This New Brain-Like Chip Could Slash AI Energy Use by 70% Apr. 23, 2026 — A breakthrough in brain-inspired computing could make today’s energy-hungry AI systems far more efficient. Researchers have engineered a new nanoelectronic device using a modified form of hafnium ... Scientists May Have Found the Holy Grail of Quantum Computing Feb. 21, 2026 — Scientists may have spotted a long-sought triplet superconductor — a material that can transmit both electricity and electron spin with zero resistance. That ability could dramatically stabilize ... The Secret to Human Intelligence? It Might Be in Our Gut Jan. 5, 2026 — New research shows gut bacteria can directly influence how the brain develops and functions. When scientists transferred microbes from different primates into mice, the animals’ brains began to ... Showing AI Users Diversity in Training Data Boosts Perceived Fairness and Trust Oct. 22, 2024 — While artificial intelligence (AI) systems, such as home assistants, search engines or large language models like ChatGPT, may seem nearly omniscient, their outputs are only as good as the data on ... Scientists Claim That Overeating Is Not the Primary Cause of Obesity Sep. 13, 2021 — A perspective article challenges the 'energy balance model,' which says weight gain occurs because individuals consume more energy than they expend. According to the authors, ... TRENDING AT SCITECHDAILY.com Astronomers Find a New Clue for Detecting Runaway Supermassive Black Holes Why You Hate Eating Bugs: DNA Reveals a 9,000-Year-Old Legacy Coffee May Protect the Liver in More Ways Than Scientists Realized The Hidden Health Risk of Sitting for Hours Without a Break

Source: ScienceDaily AI — Published January 5, 2026 at 00:08 UTC — Category: Research