**BOSTON, MA & LOS ANGELES, CA** – Groundbreaking advancements in artificial intelligence and materials science are poised to redefine technological efficiency and sustainability. Researchers at Tufts University have unveiled a hybrid AI system that dramatically cuts energy consumption by up to 100 times while simultaneously boosting accuracy in robotic tasks. Concurrently, a team at the University of California, Los Angeles (UCLA) has developed a novel method for synthesizing MXene materials using molten salts and iodine, achieving a remarkable 160-fold increase in their electrical conductivity. These breakthroughs, announced in late March and early April 2026, signal a significant leap forward in addressing critical energy demands and enhancing material performance across diverse applications.

The artificial intelligence innovation, termed neuro-symbolic AI, combines the strengths of traditional neural networks with human-like symbolic reasoning. Developed in the laboratory of Matthias Scheutz, Karol Family Applied Technology Professor at Tufts University, this approach guides decision-making with logical rules rather than solely relying on statistical pattern matching, akin to how humans break down complex problems into structured steps. In rigorous tests, including the classic "Tower of Hanoi" puzzle, the neuro-symbolic system achieved an impressive 95% success rate, significantly outperforming conventional models that managed only 34%. Furthermore, the system drastically reduced training time from over a day and a half to just 34 minutes, consuming only 1% of the energy required by standard visual-language-action (VLA) models, with operational energy use also dropping to just 5%.

This surge in AI efficiency arrives as the technology's energy footprint garners increasing scrutiny. According to the International Energy Agency, AI systems and data centers in the United States consumed approximately 415 terawatt-hours of electricity in 2024, accounting for over 10% of the nation's total energy output, a figure projected to double by 2030. Such rapid growth has raised substantial sustainability concerns, with current large language models (LLMs) and VLAs often being resource-intensive and prone to producing unreliable outputs or "hallucinations". The Tufts breakthrough offers a more sustainable and dependable foundation for future AI systems, potentially easing the strain on electrical grids and reducing the need for new fossil-fuel-powered data center infrastructure.

In parallel, a major materials science advancement has transformed MXenes, a class of ultra-thin, two-dimensional materials known for their exceptional electrical conductivity and energy storage capabilities. Led by Dr. Xiaofeng Qian at UCLA, the new synthesis method utilizes molten salts and iodine vapor to create MXenes with perfect atomic order, boosting their conductivity by up to 160 times compared to previous iterations. This innovative "GLS method" circumvents the harsh chemical etching processes traditionally used, which often leave MXene surfaces disordered and limit performance. "Our new method enables the creation of MXenes with perfect atomic order," stated Dr. Qian. Dr. Mahdi Ghorbani-Asl from Helmholtz-Zentrum Dresden-Rossendorf (HZDR) emphasized the importance of surface atoms, explaining they "strongly influence how electrons move through the material, how stable it is, and how it interacts with light, heat, and chemical environments".

MXenes, first discovered in 2011 by Drexel University researchers, have shown great promise in various applications, including energy storage, electromagnetic interference shielding, and biosensing. However, their widespread adoption has been hindered by challenging, imprecise production methods that often involve hazardous chemicals like hydrofluoric acid (HF). The atomic disorder resulting from these older techniques trapped and scattered electrons, much like "potholes slowing traffic on a highway," according to Dr. Dongqi Li from TU Dresden. The new molten salt and iodine method offers a cleaner, more controlled alternative, allowing precise control over surface halogen atoms and reducing unwanted impurities.

The implications of these dual breakthroughs are profound. The neuro-symbolic AI approach promises to make AI development more accessible and environmentally responsible, paving the way for advanced AI capabilities without an unchecked increase in energy demand. For MXenes, the enhanced conductivity and controlled synthesis open doors for the development of next-generation energy storage devices, such as more efficient supercapacitors and batteries. The ability to fine-tune surface composition also enables customization for advanced electronics, catalysis, photonics, and a host of other applications. These scientific leaps collectively point towards a future where technological progress is more sustainable, efficient, and reliable.