Anthropic's Claude Was Weaponized by Spies, Hackers, and Propaganda Farms
Anthropic's latest threat report details how state actors, criminals, and hacktivists weaponized Claude across cyber, surveillance, biology, and distillation campaigns.
20 stories
Anthropic's latest threat report details how state actors, criminals, and hacktivists weaponized Claude across cyber, surveillance, biology, and distillation campaigns.
A new recurrent reasoning method mixes denoising with looped hidden states, hitting 58.8% on ARC-AGI-1 and 12.2% on ARC-AGI-2 with just 7M parameters.
Edge0 is an open source framework that streams MoE experts from SSD, letting a 35B parameter model run on a 24GB Mac mini with under 3GB active memory.
Google folded the full Gemini API documentation directly into AI Studio, so developers can read reference material without leaving the build surface.
Cohere's new 218B MoE translation model tops WMT26 against DeepL, Google Translate, and open alternatives across 50+ languages under a non-commercial license.
Google launches a native Gemini desktop app for Windows 10 and 11 with an Alt+Space hotkey, Spark agent access, and image and video generation.
Google Pics packages Nano Banana into a Workspace-native image editor with object-level control, in-image text editing, and live collaboration.
Cognition's new coding model matches frontier systems while costing up to 70% less, driven by an RL recipe that trains every effort level at once.
Black Forest Labs shipped a prompt-driven video editor that changes one thing per clip while leaving length, framing, and audio untouched.
Krea's new agent platform reads your canvas, plans multi-model pipelines, and executes creative jobs from a single prompt, no node-wiring required.
Tencent Hunyuan opens the code and weights for a 1.5B unified speech model that handles TTS, editing, denoising, and separation from plain instructions.
Unitree open-sourced a 6B-parameter humanoid foundation model that runs 64 tasks with one network across grippers, dexterous hands, and full-body motion.
DeepSeek's 552B mixture-of-experts model splits input and output compute asymmetrically, natively handles vision, and undercuts its own flagship on price.

Google's Pixel Test Engineering team open-sourced ARTEMIS, an Android automation agent hitting 99%+ on AndroidWorld with 3-5s step latency.
Together AI highlights that Moonshot's open-weight Kimi K3 outscores Anthropic's newer Claude Fable 5.1 by 60% on the hard slice of Harvey's autonomous legal agent benchmark.
A new open-source agent skill called dream-loop turns any capable coding agent into a self-critiquing 3D artist that iterates against generated concept art.
Zed's weekly release adds call hierarchy navigation, multi-select Git staging, auto language detection for scratch buffers, and live search-as-you-type by default.
CozyClay turns your browser into a previs studio where you block shots, pose characters, and hand the same camera moves to an AI video model.
Epoch AI's new explorer estimates compute capacity for OpenAI, Google DeepMind, Anthropic, Meta, and xAI, revealing a 17x surge at OpenAI.
A new evolutionary model shows that when computation, replication, and social behavior all draw from one energy budget, cooperation emerges naturally in populations of random Z80 programs.