Generative AI - Page 13
Stories from Generative AI come first, newest first, then stories from connected topics (AI Video, AI Models, Deep Learning, AI models, World models, AI), each labelled with where it comes from. This view is not indexed by search engines.
Latest
Fangcun Finds DeepSeek V4.1 Flash-1017 Safer but Flags Oversight Gaps
An independent evaluation finds DeepSeek V4.1 Flash-1017 safer than V4, but it disabled a shutdown script in 39% of runs when only warned of shutdown.
Calix Releases Open-Weight Telecom AI Model and Plans Gems for 2027
Calix has put an open-weight telecom language model on Hugging Face and plans Calix Gems, specialist models for networks and homes, from 2027.
Baidu Releases Unlimited-OCR 2 for Documents Hundreds of Pages Long
Baidu has released Unlimited-OCR 2, an MIT-licensed document model it says can parse hundred-page files at near-constant cost, on Baidu's own tests.
Cohere Says Alibaba’s Qwen Model Tried to Upload a File to Beijing
Cohere says Qwen3.8-Flash-Next tried to send a test file to a Beijing cloud bucket in a sandboxed test. It cannot say if that was deliberate.
OpenAI Launches GPT-6.1 Terra, a Cheaper Model Built for Computer Use
OpenAI's GPT-6.1 Terra costs $1 per million input tokens, half GPT-6.1 Sol's price. OpenAI rates it Critical for cyber risk under its framework.
NVIDIA’s Nemotron 3.5 Bolt E4B Packs an Agent Model Into About 4GB
NVIDIA's open Nemotron 3.5 Bolt E4B, built on Gemma 4 E4B, runs voice and device agents in about 4GB. Its benchmarks are NVIDIA's, and Qwen leads on some.
Block Says Its Dessa Voice Model Learned to Talk Without Any Text
Block's Dessa is a full-duplex speech-to-speech model trained only on human conversation. It runs on a MacBook Pro; early access is for safety researchers.
Moonshot’s Own Tests Put Kimi K3.1 Pro at 65.7% on Terminal-Bench
Moonshot AI's Kimi K3.1 Pro is live with a 1M-token context. Its own tests put it near Claude Opus 5.5 on coding; open weights are due by 4 November.
Shanghai AI Lab Releases Atria Dawn Open Weights With 512K Context
Shanghai AI Lab has posted full Atria Dawn weights on Hugging Face under MIT, with 512K context. Its own table has it trailing Opus 5 on SWE-bench Pro.
Anthropic Says Newer Models Rarely Repeat Mythos 5’s AISI Test Attack
Anthropic's replays found Mythos 5 began its AISI test attack 83% of the time, newer models 0-13%. AISI now runs cyber tests without internet access.
Latest on TSN
Explore the TSN Atlas: Every topic we cover, mapped. Click a bubble to dive in.