This week's list of top data news covers December 13, 2025 to December 19, 2025, and includes articles about robots that build physical objects from text descriptions and deep learning models that predict how a cell's shape will change.
1. Prediction of cell behavior during early development
MIT researchers have developed a deep learning model that predicts how individual cells move, fold, divide, and rearrange during the early stages of fruit fly development. The model analyzes high-resolution videos of embryos and learns geometric patterns such as cell location, shape, and contacts with neighboring cells. Using this information, we can predict minute-by-minute cell behavior with approximately 90% accuracy during the first hour of development. The researchers now plan to apply this approach to other species and human tissues to uncover early disease patterns, including those associated with asthma and cancer.
2. Understanding the solar surface with AI
Researchers at the University of Hawaii Astronomy Institute have developed an AI tool that creates accurate 3D maps of the Sun's magnetic field. The tool begins with observations by the Daniel K. Inouye Solar Telescope, which measures the magnetic field in the Sun's atmosphere, but uncertainties remain about its direction and height. AI tools can help bridge the gap by combining these observations with the fundamental physical laws that govern the behavior of magnetic fields. This will allow scientists to determine the true direction and height of magnetic structures, allowing them to more accurately predict solar activity that may affect Earth.
3. Describe objects and construct objects
Researchers at MIT and partner institutions have developed an AI-powered robotic system that can build multi-component objects, such as chairs and shelves, directly from simple text instructions. The system translates user prompts into 3D designs, explaining the reasoning behind the object's shape and use, and determining how prefabricated parts should be placed. A robotic arm then physically assembles the object from reusable components. This system makes physical design faster, more intuitive, and more accessible.
4. Improving the quality of Qubit
Princeton engineers have developed a superconducting qubit that stores quantum information much longer than current designs by addressing one of the core limitations of quantum computing: energy loss caused by imperfect materials. By using tantalum metal, which can withstand aggressive manufacturing without introducing defects, and combining it with ultra-high quality silicon, the researchers minimized the microscopic defects that typically disrupt qubit stability. As a result, qubits maintain coherence (the ability to retain information) up to 15 times longer than many existing processors.
5. Amplify your audio with Meta's AI glasses
Meta has begun rolling out a software update for its AI-powered smart glasses that adds a new audio mode focused on conversation. In noisy environments such as restaurants or bars, the glasses are designed to use AI and a microphone to pick up sounds from in front of you, rather than surrounding noise, and amplify the voices of the people the wearer is listening to. The AI system processes incoming audio in real-time, identifying speech patterns and filtering out competing sounds to keep the intended speaker's voice clear and easy to hear.
6. Modeling materials in superhypersonic flight
Scientists at Sandia National Laboratories have developed a computational model that predicts how protective materials that prevent overheating in high-speed vehicles will behave during hypersonic flight. The model learns from laboratory tests that show how the material heats and wears, as well as flight data that shows how the material behaves in real-world high-speed conditions. By linking these two data sets, the model can estimate how the material will perform before it flies. This allows engineers to quickly compare and improve thermal shield designs.
7. Truck tire failure detection
The Utah Department of Transportation has installed an AI-powered roadside tire monitoring system at truck drop-off locations to detect tire issues. The system uses in-road sensors and machine learning to identify abnormal patterns, such as uneven pressure distribution or unusual vibrations, and alert inspectors to take a closer look at the tire during routine inspections. UDOT plans to expand this technology to all ports of entry to enhance highway safety and reduce avoidable accidents involving heavy cargo vehicles.
8. Convert live sports footage to 3D game data
Peripheral Labs uses AI and computer vision to transform standard sports videos into 3D digital views of games. The system analyzes footage to understand where players are on the field, how they are moving, and what their body position is at each moment. This allows you to observe the play from all angles, follow a single player, or pause to examine key moments.
9. Make smart home devices work across brands
IKEA has launched a new line of smart home products built around Matter, a shared smart home standard that allows devices from different brands to work together. Matter provides a common communication system so smart lights, sensors, plugs, and remote controls can exchange information instead of operating in separate apps or closed systems. Each device sends simple signals, such as detecting movement, turning on lights, or using energy, to a shared network where other devices can respond. For IKEA, this makes smart home products easier to combine with non-IKEA devices, expanding their usefulness and range.
10. Turn emergency calls into medical records
A regional emergency medical service in the south of England is testing an AI tool within its emergency call centre. At these call centers, clinicians speak with patients over the phone, assess their symptoms, and determine what care they need. AI software records these phone conversations and automatically converts them into structured medical notes that clinicians can review and approve. The goal is to reduce time spent on post-call documentation, allowing clinicians to see more patients and focus on decision-making rather than paperwork.
