UC San Diego and MIT researchers have published a landmark study in Science demonstrating a scalable method to steer and monitor AI models by directly manipulating internal concept representations, exposing both safety vulnerabilities and capability improvements.

MIT researchers develop language model for codon optimization, boosting protein production including trastuzumab by 25-300%, published in PNAS.

MIT Professor Jim Collins leads groundbreaking research using generative AI to design programmable antibacterials targeting drug-resistant pathogens.

MIT researchers develop groundbreaking AI-powered software that automatically segments eight distinct nerve fiber bundles in brainstem MRI scans.

MIT CSAIL introduces EnCompass framework enabling AI agents to backtrack and optimize LLM outputs, achieving 15-40% accuracy boost with 82% less code.

MIT engineers develop DiffSyn, a diffusion-based generative AI model trained on 23,000 synthesis recipes that suggests promising pathways for creating new materials in under a minute, dramatically accelerating experimentation and discovery timelines.

New MIT interdisciplinary course examines rationality in AI systems, blending computer science and philosophy for next-gen scholars.

MIT researchers demonstrate that best-performing machine learning models can become worst-performing when applied to new data environments, revealing hidden risks from spurious correlations in medical AI and other critical applications.