Full-text search across 26947 articles. Combine with topic and date filters; results sorted by relevance.
Eighty-eight percent of enterprises have reported at least one security incident tied to AI agents, and 61% of those trace back to over-permissioned credentials, according to a report cited in a new article that outlines…
Meta released Muse Glimmer, a 30-billion-parameter AI model designed to run locally on a Mac or PC with a single consumer GPU, with weights available for download from Hugging Face under an Apache 2.0 license. The model,…
The FCC is preparing restrictions on Chinese optical transceivers, causing shares of Zhongji Innolight and Eoptolink to briefly fall nearly 10% in a single day, though Coherent's factory in Wuxi complicates the supply ch…
Enterprise AI inference costs hit a yearly low of $1.16–$1.18 per million tokens from August 6–8, down from $2.04 on May 31 and $1.45 in late July, according to data from investment bank Jefferies and research firm Silic…
Meta CEO Mark Zuckerberg pushed back against predictions of an AI-driven job apocalypse, arguing in an essay posted Monday that AI will create an 'abundance of jobs' and new careers such as world builders and personal bi…
Kimi K3, a ~2.8T parameter open-weight coding model from Moonshot AI, ranks first on Frontend Code Arena with a score of 1,679, ahead of the closed Fable 5 at 1,631, and its weights were released on Hugging Face on 2026-…
Microsoft plans to unveil its Maia 300 artificial intelligence chip as soon as September, according to The Information, citing people with direct knowledge. The company has talked with Taiwan Semiconductor Manufacturing …
Docker Inc. launched Docker Sandboxes, a new product that isolates AI coding agents in microVMs with their own kernels and private Docker daemons, rather than containers, to make bypass-mode agents safe by default. The c…
Microsoft Corp. plans to significantly increase production of its next-generation AI chips, the Information reported, citing people familiar with the situation. The company is in discussions with Taiwan Semiconductor Man…
OpenAI is slowing work on its unreleased Astra model after internal evaluations indicated it could not rule out a 'Critical' cybersecurity capability rating under its Preparedness Framework, prompting expanded safety tes…
Enterprise AI inference costs hit a 2026 low, with average prices ranging between US$1.16 and US$1.18 per million tokens from August 6 to 8, according to investment bank Jefferies, citing data from US research firm Silic…
Meta launched Muse Glimmer, a 30-billion-parameter open-weight AI model designed for autonomous agentic tasks, capable of running locally on consumer hardware with a context window exceeding 131,000 tokens and quantized …
Corma, a Tel Aviv and San Francisco-based startup founded in 2025, emerged from stealth with $60 million in seed funding led by Sequoia Capital, with participation from Khosla Ventures and Coatue, to build AI models for …
Meta released Muse Glimmer, a 30-billion-parameter open-weight small language model available on Hugging Face under the Apache 2.0 license, designed to run locally on consumer hardware with a single GPU. Developed by Met…
Microsoft's custom AI chip program is back on track, with the Maia 200 inference chip now deploying in data centers in Iowa and Arizona and the Maia 300 already in design. The Maia 200, built on TSMC's 3nm process, offer…
OPSWAT's AI Content Inspector adds a content authenticity verdict to its MetaDefender pipeline to detect AI-generated invoice fraud, which produces convincing fake invoices that pass standard malware and PDF security sca…
The Institute for Progress (IFP) published 23 policy recommendations across seven categories to help policymakers address risks from increasingly automated AI research and development, including transparency, risk manage…
DeepSeek, the Chinese AI lab, released an updated V4-Flash model that delivers better performance at a lower price than its V4-Pro, according to a The New Stack analysis. The V4-Flash costs $0.28 per million input tokens…
Buzz, a channel-based chat workspace, now allows users to configure autonomous agents as playable characters with names, cryptographic identities, and personalities. Agents are reactive, responding to @mentions, and can …
A developer's review of recent AI releases found only two updates met a strict 36-hour cutoff: Meta's Muse Glimmer, a 30B multimodal model under Apache 2.0, and Hugging Face's Transformers 5.15.0. The developer emphasize…