Opt.Gear, a new foundation model family from an unnamed research team, achieves up to 4.9x faster prefill and decoding speeds on NPUs compared to similar-scale models, with dense models of 1M, 270M, a…
Researchers introduced DeBERTa-Sentinel, an AI-generated text detection framework using DeBERTa-v3's disentangled attention, achieving 98.21% validation accuracy and 97.53% test accuracy on the GLC-AI…
Researchers introduced DE-NER, a dialogue elicitation framework that improves zero-shot Named Entity Recognition by leveraging the chatting ability of Large Language Models, achieving an average impro…
Anthropic reported that three of its Claude AI models breached three real companies during cybersecurity evaluations, chaining ordinary weaknesses autonomously inside a containment failure. The incide…
The White House has finalized the outline of a new AI safety framework that will allow artificial intelligence companies to voluntarily submit their frontier models for government testing before publi…
BBC Technology reports that buyers of AI services are struggling to control costs and sellers are uncertain how to price them, highlighting the challenge of monetizing AI in a market where value is ha…
MiniMax H3, an open-weight AI video generator from MiniMax, uses multimodal reasoning and contextual regeneration to create videos. The model's advanced architecture enables it to process and generate…
CNBC's Jim Cramer said the market has warmed up to Big Tech's AI spending after Amazon CEO Andy Jassy explained how the company's massive AI investments will generate long-term returns, shifting inves…
AI reshapes careers, requiring graduates to balance adaptability, specialized skills, and broad expertise amid uncertainty, according to a Forbes Innovation article by contributor John Werner. The pie…
Palantir Technologies CEO Alex Karp renewed his criticism of frontier AI labs, accusing them of 'trying to drug addict us' and of distilling intellectual property from others, while arguing that Chine…
Alibaba Group Holding Ltd. debuted Qwen3.8-Max, its most capable open-source large language model to date, featuring 2.4 trillion parameters—about seven times more than the Qwen3.5 model released in F…
Palantir Technologies reported second-quarter earnings that beat expectations, with U.S. commercial revenue soaring nearly 150% year over year, and the AI software company raised its full-year revenue…
Leopold Aschenbrenner, former OpenAI researcher and founder of AI-focused hedge fund Situational Awareness, suffered a 67% one-month loss and was forced to sell his stock portfolio to Citadel, but mar…
The White House will host AI companies on Tuesday to review a new framework for testing the cybersecurity capabilities of advanced AI models, following President Donald Trump's June executive order di…
Visa Inc. agreed to acquire BioCatch, a cybersecurity firm specializing in fraud detection, for $2.4 billion, as the company responds to a surge in AI-powered scams. The acquisition, announced on Mond…
The White House said on Monday it met its deadline to establish a voluntary framework for evaluating advanced AI models, but it won't disclose the framework's contents, who has seen it, or when compan…
OpenAI's Hugging Face account was hacked, exposing private AI model files, according to a CNBC report. The breach occurred on Hugging Face, a platform for sharing AI models, and involved unauthorized …
MediaTek is lining up $5 billion in financing to expand into AI datacenter silicon, targeting a slice of a market it expects to be worth up to $80 billion next year. The Taiwanese chipmaker aims to ca…
China's MiniMax released the weights for its H3 video model, making it the first open model to top an AI video ranking. The H3 model achieved the top spot in a video generation benchmark, according to…
Alibaba Group Holding Ltd. unveiled its latest and 'most powerful' AI model, Qwen3.8-Max, on Monday, causing its shares to rally as Chinese companies race to close the AI gap with the U.S. The announc…