Large Language Models News

Chinese AI labs are reportedly closing the gap with U.S. AI capabilities by developing more efficient methods for training models, despite having less access to

The Trump administration, through the Justice Department, has sided with OpenAI in The New York Times' copyright case, arguing that the public benefits of train

Anthropic has launched its new large language models, Claude Fable 5.1 and Mythos 5.1, offering significant performance gains and a 75% cost reduction for Fable

A new study by Google Research and Technion reveals that large language models (LLMs) often possess facts but struggle with recall, leading to hallucinations. T

The Authors Guild warns that unlicensed and unrestricted AI training, particularly using books, could destroy the publishing ecosystem, citing concerns over AI

AI startup Hugging Face, based in New York, is reportedly exploring a sale that could value the company at $13 billion. The company, known for hosting open-sour

Ramp has introduced its new AI model routing service named Router, which enables users and companies to utilize and switch between various large language models

AI is transforming into a financial engineering business, shifting competitive advantage from models to the balance sheets funding their infrastructure. Big Tec

A Nature review examines the rapid adoption of large language models (LLMs) in clinical care, highlighting emerging security and safety risks. The review outlin

A growing 'poison AI' movement aims to corrupt AI models like ChatGPT and Gemini by feeding them bad information, making them less useful, but this strategy car

Ally Financial's CMO, Andrea Brimmer, is leading a strategy to ensure the company is favorably recommended by AI search, emphasizing that brand reputation is in

Singaporean startup Acrab has launched the Agent Box, a mini PC powered by its proprietary G≡LIX 1 chip, designed to run 100 billion parameter language models l

AI companies are reportedly destroying rare physical books published before 2022 by spine-cutting them for rapid digitization to train large language models, ra

Artificial intelligence has achieved a new milestone by scoring a perfect 100 at the International Mathematical Olympiad, with systems from Huawei and Xiaohongs

The artificial intelligence industry's rapid innovation is raising national security concerns as generative models can be weaponized for cyber operations and su