Facebook Research unveils Tuna-2, a groundbreaking multimodal AI model that ditches traditional vision encoders in favor of direct pixel patch processing, outperforming predecessors on diverse benchmarks while supporting both image understanding and generation tasks in 7B and 2B parameter sizes.
Hugging Face CEO Clem Delangue warns that anti-open-source lobbying in the US risks handing AI dominance to a handful of corporations, as the global developer base is projected to surge from millions to 100 million builders and AI agents may soon outnumber human users on the platform.
Anthropic is racing to close a massive $50 billion funding round within two weeks at a valuation exceeding $900 billion, surpassing OpenAI's $852 billion valuation, as the AI powerhouse reports over $30 billion in annual revenue and prepares for a highly anticipated IPO.
Anthropic is eyeing a staggering $900 billion valuation in a new funding round, potentially overtaking OpenAI as the world's most valuable AI startup, while also revealing a powerful but dangerous new AI model called Mythos capable of exploiting software vulnerabilities — deemed too risky for public release after unauthorized users …
Chip Huyen's GitHub repository 'aie-book,' supporting her 2025 O'Reilly book 'AI Engineering,' surpasses 15,400 stars and 2,200 forks, offering developers a comprehensive hub covering prompt engineering, RAG, agent building, finetuning, and hallucination mitigation for foundation model development.
A groundbreaking Harvard study published in Science finds AI outperforms doctors in emergency triage with 67% diagnostic accuracy versus 50-55% for physicians, and dominates long-term treatment planning at 89% accuracy compared to just 34% for human doctors, though researchers caution that unresolved safety and accountability concerns mean AI is not …
OpenAI's o1 AI model matches or outperforms ER doctors in triage, diagnosis, and clinical case management across 76 real emergency cases, according to a Harvard-led study published in Science, with researchers urging urgent clinical trials to explore how AI can partner with physicians to reduce errors and improve patient outcomes.
MIT, WPI, and Google researchers unveil WRING, a groundbreaking debiasing technique that eliminates bias in AI vision-language models by rotating bias-responsible coordinates in high-dimensional space, avoiding the 'Whac-A-Mole dilemma' of creating new biases while fixing old ones.
Google DeepMind launches ProEval, a free open-source tool that slashes generative AI evaluation costs by up to 100x using Bayesian Quadrature techniques, achieving ±1% accuracy with a fraction of typical samples while proactively identifying model failure patterns across major benchmarks.
A groundbreaking new AI reasoning framework called LaDiR is outperforming all existing models across math and planning benchmarks by combining latent diffusion models with continuous representations to generate more accurate, diverse, and interpretable reasoning trajectories.