Listen to today’s podcast: https://www.youtube.com/channel/UC-nqwUyvLDEvs7bV985k-gQ
AI Daily Podcast 11/23/2025
Today’s podcast episode was created from the following stories: a rapid-fire look at breakthroughs in generative AI, robotics, operating system features, and the infrastructure and energy needed to power it all.
CEO Sundar Pichai’s Cheeseburger Flex Sums Up Google’s Amazing Comeback
By Alistair Barr — November 22, 2025
Google’s Gemini 3 and the upgraded Nano Banana Pro are delivering sharper spatial reasoning and more realistic image generation, helping the company reclaim AI leadership and lift its market value. The viral “cheeseburger test” highlights a deeper shift: models are getting better at understanding 3D relationships, a capability that can translate into safer, smarter real-world decision-making in areas like design and engineering. The takeaway: scaling laws still matter—and Google is executing on them.
300,000 character poses in an AI instant: my visit to Disney was a peek at animation’s new reality
By Corinne Reichert — November 22, 2025
Disney-backed startup Animaj is using AI to accelerate animation by handling motion in-betweening while keeping artists firmly in control of key poses and style. Trained on show-specific pose libraries (300,000+ poses from Pocoyo), the system slashes production time from months to weeks without sacrificing brand consistency. For studios racing to meet streaming demand, it’s a practical path to faster output that preserves the animator’s creative role.
Confused by Windows 11’s new AI actions? Here’s what they do and how to manage them
By Mauro Huculak — November 22, 2025
Windows 11 now adds “AI actions” to File Explorer’s right‑click menu, offering shortcuts like background removal, object erase, visual search, and document summaries—each handled by apps like Paint, Photos, Edge, or Microsoft 365. Users can toggle these per-action in Settings (rolling out since the September 2025 update; not yet in the EU), giving control over how much AI shows up in daily workflows. It’s a small feature with big convenience for quick edits and previews.
Why DETRs are replacing YOLOs for real-time object detection
By Larsen D’hiet — November 21, 2025
Detection Transformers (DETRs) have matured into real-time detectors that beat YOLO on accuracy at comparable speeds, thanks to advances like deformable attention, denoising, and improved query selection. Models such as D‑Fine and RF‑DETR show strong results on COCO, with the added advantage of permissive Apache 2.0 licensing that’s friendlier for commercial use. Bottom line: DETRs are now the practical choice for production systems that need speed, accuracy, and fewer licensing hurdles.
ChatGPT wants to join your group chats
By Katie Notopoulos — November 22, 2025
OpenAI rolled out group chats in ChatGPT across Free, Go, Plus, and Pro, letting the bot chime in on plans, research, and brainstorming alongside multiple people. Early testing shows utility for tasks like itineraries and recommendations, but the assistant can over-participate—highlighting the balance needed between helpful and intrusive. Expect stronger use cases in study groups and collaborative work rather than casual friend chats.
Meta se ha convertido en comercializadora de energía con un objetivo: que Estados Unidos prepare sus centrales para la IA
By Eva R. de Luis — November 23, 2025
Meta is entering U.S. electricity trading to secure long-term power contracts that can catalyze new generation capacity for its AI ambitions. The move aims to manage risk and ensure supply for energy-hungry data centers, though it brings exposure to volatile power markets. It’s a sign that Big Tech won’t just consume energy—they’ll actively shape how it’s financed and built.
Elon Musk says the job of the future is no longer coding, it’s ‘no job’ at all. Here’s how investors can prepare.
By Barchart Insights — November 21, 2024
Musk argues AI will make most jobs optional, shifting human value from execution to direction and judgment. The investment lens favors AI infrastructure (GPUs and servers), human–AI collaboration platforms, automation leaders, and cybersecurity as adoption expands. The core thesis: AI does the work; humans set the purpose—and portfolios should reflect that divide.
This startup bucked mainstream robotics training methods to teach a robot how to load the dishes
By Lloyd Lee — November 22, 2025
Sunday Robotics unveiled “Memo,” a home robot that autonomously clears tables and loads dishwashers, trained via custom gloves that mirror the robot’s hands—avoiding costly teleoperation. The approach claims two orders of magnitude better capital efficiency and scales through distributed data collection. Demonstrations of delicate manipulation (like handling wine glasses) hint at meaningful progress in practical home robotics.
Together, these stories chart AI’s sprint from labs to living rooms: smarter models powering better tools, robots tackling real chores, and platforms weaving AI into everyday collaboration. They also spotlight the real-world requirements—compute, policy, and power—to sustain that momentum. As AI accelerates, the edge will belong to those who pair human judgment with the right infrastructure and guardrails.
