A Brazilian prompt engineer achieves a mere 7.5% score on a challenging new AI coding test, underscoring the complexity of real-world coding tasks for artificial intelligence models and prompting the organizer to offer a $1 million prize for an open-source model exceeding 90%.
OpenAI unveils GPT-5, a groundbreaking language model set for August release, boasting reasoning prowess from the o3 model, with CEO Sam Altman teasing its impressive capabilities.
Amid rising U.S.-China tensions and recent layoffs at Amazon Web Services, the tech giant shuts down its AI research lab in Shanghai as part of strategic adjustments and reassessment of operations in China.
Aeneas, a groundbreaking AI model, revolutionizes the interpretation of ancient Latin inscriptions by reasoning across thousands of texts, retrieving contextual parallels to restore fragments, predict dates and locations, and attribute inscriptions to Roman provinces, transforming historians' ability to connect the past.
A UK engineering researcher spearheads an NSF-funded initiative, collaborating with Michigan State University, to develop cutting-edge machine learning models capable of simulating the progression of heart fibrosis, advancing computational modeling of heart disease.
In a coding marathon showdown at the AtCoder World Tour Finals, Polish programmer Przemysław Dębiak triumphs over OpenAI's AI model, outscoring it by 9.5% with an innovative heuristic-driven approach that showcases human creativity's edge over AI's raw speed.
In a groundbreaking achievement, the AI system Gemini, powered by Deep Think, conquers the prestigious International Mathematical Olympiad, earning a gold medal with a perfect score on five out of six complex math problems through innovative parallel thinking and reinforcement learning techniques.
Groundbreaking AI models from OpenAI and Google DeepMind achieve gold medal-level performance at the prestigious International Mathematical Olympiad, solving complex problems and providing natural language proofs, marking a milestone in general-purpose AI for math and reasoning.
Meta unveils a groundbreaking AI framework, MTTT, that enables models to learn how to learn at test time, removing the need for human-designed self-supervised tasks and outperforming linear attention on long sequences while maintaining linear complexity.
OpenAI researchers Jason Wei and Hyung Won Chung depart for Meta's superintelligence lab, amid an AI talent war where Meta offers up to $300 million over four years to top AI talent.