Mantic's AI prediction engine beats human forecaster average in Metaculus Fall Cup, marking breakthrough in AI's ability to predict real-world events.

Researchers at University of Michigan developed Prima, an AI system that interprets brain MRI scans in seconds with 97.5% accuracy, automatically flagging emergencies.

OpenAI releases GPT-5.3-Codex, the most capable agentic coding model to date. The groundbreaking AI helped build and deploy itself, runs 25% faster.

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

Discovery Learning method enables rapid battery lifetime prediction in one week versus traditional months-long testing cycles.

Impulse AI unveils autonomous ML platform that ranked top 2.5% on Kaggle competition, automating entire workflow from data prep to production deployment in under one hour.

New AI safety report warns of proliferating deepfakes, AI companions, and autonomous systems while highlighting gold-medal AI performance in mathematics.

Comprehensive analysis identifies five recurring pitfalls driving 85% ML project failure rate: wrong problem selection, data quality issues, model-to-product gap, offline-online mismatch, and non-technical blockers, with actionable solutions for practitioners.

From LangChain to Hugging Face Transformers, these 16 open-source projects are providing the foundational tools and frameworks that are accelerating innovation in AI and machine learning.

Artificial intelligence is expanding its reach into space exploration, with researchers from Stanford University successfully implementing machine learning on robots aboard the International Space Station. The AI system has improved the efficiency of robot movement planning by 50-60%, showcasing the potential of AI to drive new opportunities in space.

Researchers have developed a new AI method called Riff-Diff that transforms enzyme design, creating highly efficient and stable biocatalysts for industrial and medical applications. The findings were published in the journal Nature.

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.

ScienceDaily reports researchers used machine learning to analyze cancer data from 185 countries, identifying specific policy changes that could improve survival rates in each nation.

Emerging world models technology aims to solve AI consistency issues by giving machines better understanding of space and time.