https://shshell.com/blog/google-gemini-omni-1-2-flash-long-clips ShShell AI Research en 2026-11-08T00:00:00.000Z Gemini Omni 1.2 Flash Shows Why Long-Video AI Is a Retrieval Problem https://shshell.com/blog/nvidia-ai-semiconductor-fabs ShShell AI Research en 2026-11-08T00:00:00.000Z AI Is Changing Semiconductor Fabs, and NVIDIA’s Factory Story Has a Measurement Problem https://shshell.com/blog/nvidia-open-source-6g-ai-tools ShShell AI Research en 2026-11-08T00:00:00.000Z NVIDIA’s Open-Source 6G Push Makes Wireless Research an AI Systems Test https://shshell.com/blog/nvidia-vera-cpu-agent-throughput ShShell AI Research en 2026-11-08T00:00:00.000Z NVIDIA’s Vera CPU Bet Says Agent Speed Depends on the Host, Not Just the GPU https://shshell.com/blog/openai-ai-agents-wet-lab-science ShShell AI Research en 2026-11-08T00:00:00.000Z OpenAI’s Wet-Lab Agent Test Moves AI Science Claims From Answers to Experiments https://shshell.com/blog/cyberarena-live-agent-attack-defense ShShell AI Research en 2026-11-07T13:00:00.000Z CyberArena Turns AI Security From a Puzzle Into a Live Attack-and-Defense Match https://shshell.com/blog/lerobot-v070-robot-learning-stack ShShell AI Research en 2026-11-07T13:00:00.000Z LeRobot v0.7.0 Pushes Open Robot Learning Toward a Reusable Data-and-Policy Stack https://shshell.com/blog/nvidia-nemotron-transcribe-granary-v2 ShShell AI Research en 2026-11-07T13:00:00.000Z NVIDIA’s Nemotron Transcribe and Granary v2 Make Speech AI a Promptable Data Pipeline https://shshell.com/blog/phonon-tts-on-device-speech-benchmarks ShShell AI Research en 2026-11-07T13:00:00.000Z Phonon’s On-Device TTS Push Shows Why Voice AI Wins or Loses on the Last Mile https://shshell.com/blog/statebench-long-document-state-tracking ShShell AI Research en 2026-11-07T13:00:00.000Z StateBench Tests the AI Skill That Summaries Keep Missing: Tracking What Changed