https://tokenstrom.com/posts/lawsuit-chatgpt-urged-woman-to-quit-epilepsy-drug-and-hide-it-from-doc/Tokenstromen2026-10-29T00:24:49+00:00Lawsuit: ChatGPT Urged Woman to Quit Epilepsy Drug and Hide It From Doctors https://tokenstrom.com/posts/opera-brings-local-ai-into-the-browser-with-its-webai-standard/Tokenstromen2026-10-29T00:23:32+00:00Opera Brings Local AI Into the Browser With Its WebAI Standard https://tokenstrom.com/posts/qwen-image-2-1-alibaba-s-image-model-adds-true-transparency-but-a-stri/Tokenstromen2026-10-29T00:22:54+00:00Qwen-Image-2.1: Alibaba's image model adds true transparency, but a stricter license https://tokenstrom.com/posts/microsoft-unveils-mai-model-family-reasoning-voice-code-and-image/Tokenstromen2026-10-29T00:22:53+00:00Microsoft unveils MAI model family: reasoning, voice, code and image https://tokenstrom.com/posts/experts-adopt-an-ai-s-hidden-bias-86-of-the-time-anthropic-study-finds/Tokenstromen2026-10-28T23:48:33+00:00Experts adopt an AI's hidden bias 86% of the time, Anthropic study finds https://tokenstrom.com/posts/unreleased-anthropic-ai-chains-bugs-in-chrome-safari-windows-and-linux/Tokenstromen2026-10-28T23:48:33+00:00Unreleased Anthropic AI chains bugs in Chrome, Safari, Windows and Linux https://tokenstrom.com/posts/anthropic-buys-one-person-biotech-that-ran-11-labs-with-claude/Tokenstromen2026-10-28T23:47:39+00:00Anthropic buys one-person biotech that ran 11 labs with Claude https://tokenstrom.com/posts/new-ai-index-open-models-lose-1216-points-mistral-large-4-drops-to-25/Tokenstromen2026-10-28T21:54:15+00:00New AI index: open models lose 12–16 points, Mistral Large 4 drops to 25 https://tokenstrom.com/posts/openai-retires-gpt-4o-and-gpt-4-1-on-jan-14-azure-and-bedrock-differ/Tokenstromen2026-10-28T21:54:15+00:00OpenAI retires GPT-4o and GPT-4.1 on Jan. 14 – Azure and Bedrock differ https://tokenstrom.com/posts/anima-ref-adapter-generate-a-character-from-one-reference-image-in-com/Tokenstromen2026-10-28T21:53:12+00:00Anima Ref Adapter: Generate a Character from One Reference Image in ComfyUI https://tokenstrom.com/posts/github-open-sources-replica-ai-prototyping-with-your-own-design-system/Tokenstromen2026-10-28T21:51:26+00:00GitHub open-sources Replica: AI prototyping with your own design system https://tokenstrom.com/posts/nvidia-s-open-nemotron-3-5-transcribe-handles-transcription-translatio/Tokenstromen2026-10-28T17:51:31+00:00Nvidia's open Nemotron 3.5 Transcribe handles transcription, translation, speakers https://tokenstrom.com/posts/gemini-4-carbon-google-s-most-powerful-model-launches-for-cyber-defend/Tokenstromen2026-10-28T17:51:13+00:00Gemini 4 Carbon: Google's most powerful model launches for cyber defenders first https://tokenstrom.com/posts/ai-agents-breach-anthropic-startup-gets-in-through-third-party-softwar/Tokenstromen2026-10-28T13:49:58+00:00AI agents breach Anthropic: Startup gets in through third-party software https://tokenstrom.com/posts/qwenputer-a-risc-v-processor-running-inside-qwen-3-5-s-weights/Tokenstromen2026-10-28T13:49:58+00:00Qwenputer: A RISC-V processor running inside Qwen 3.5's weights https://tokenstrom.com/posts/don-t-launch-it-in-downloads-tampered-agents-md-hijacks-coding-agents/Tokenstromen2026-10-28T09:50:38+00:00Don't launch it in Downloads: Tampered AGENTS.md hijacks coding agents https://tokenstrom.com/posts/clover-35b-a3b-diffusion-language-model-beats-qwen3-6-at-agent-tasks-a/Tokenstromen2026-10-28T09:49:20+00:00Clover-35B-A3B: Diffusion language model beats Qwen3.6 at agent tasks, Apache 2.0 https://tokenstrom.com/posts/us-senator-probes-openai-after-ai-agents-broke-out-and-hacked-hugging/Tokenstromen2026-10-28T05:48:43+00:00US senator probes OpenAI after AI agents broke out and hacked Hugging Face https://tokenstrom.com/posts/liquid-ai-s-lfm3-tiny-2b-models-run-in-your-browser-via-webgpu/Tokenstromen2026-10-27T21:48:21+00:00Liquid AI's LFM3: tiny 2B models run in your browser via WebGPU https://tokenstrom.com/posts/meta-releases-muse-glimmer-an-open-30b-model-you-can-run-at-home/Tokenstromen2026-10-27T21:46:36+00:00Meta releases Muse Glimmer, an open 30B model you can run at home https://tokenstrom.com/posts/anthropic-claude-models-hacked-three-real-organizations-during-testing/Tokenstromen2026-10-27T21:46:35+00:00Anthropic: Claude models hacked three real organizations during testing https://tokenstrom.com/posts/anthropic-s-ipo-filing-ai-agents-broke-out-of-a-test-environment/Tokenstromen2026-10-27T21:45:37+00:00Anthropic's IPO filing: AI agents broke out of a test environment https://tokenstrom.com/posts/openai-claims-yang-mills-proof-beating-anthropic-mathematicians-to-it/Tokenstromen2026-10-27T21:45:37+00:00OpenAI claims Yang-Mills proof, beating Anthropic mathematicians to it https://tokenstrom.com/posts/liquid-ai-s-lfm3-tiny-2b-models-for-speech-vision-and-video-land-in-ol/Tokenstromen2026-10-27T17:45:15+00:00Liquid AI LFM3: Tiny 2B models for speech, vision and video land in Ollama https://tokenstrom.com/posts/oxr-open-1-3b-document-parsing-model-tops-three-benchmarks/Tokenstromen2026-10-27T17:44:06+00:00OXR: Open 1.3B document parser tops three benchmarks, claims its maker https://tokenstrom.com/posts/axiom-math-ai-agents-deliver-first-fully-machine-checked-proof-of-fini/Tokenstromen2026-10-27T17:43:48+00:00Axiom Math: AI agents deliver first fully machine-checked proof of finite simple groups https://tokenstrom.com/posts/sensenova-6-8-code2video-open-27b-model-writes-code-that-makes-videos/Tokenstromen2026-10-27T16:52:20+00:00SenseNova 6.8 Code2Video: Open 27B model writes code that makes videos https://tokenstrom.com/posts/genie-4-deepmind-s-world-model-now-generates-worlds-with-sound-and-dia/Tokenstromen2026-10-27T16:50:47+00:00Genie 4: DeepMind's world model now generates worlds with sound and dialogue https://tokenstrom.com/posts/baidu-releases-unlimited-ocr-2-for-documents-hundreds-of-pages-long/Tokenstromen2026-10-27T08:50:21+00:00Baidu releases Unlimited OCR 2 for documents hundreds of pages long