Listen to today’s podcast: https://www.youtube.com/channel/UC-nqwUyvLDEvs7bV985k-gQ
AI Daily Podcast — December 4, 2025
Today’s podcast episode was created from the following stories:
Ex-Google CEO Eric Schmidt says AI isn’t overhyped — the biggest gains from automating corporate work are still ahead
By Thibault Spirlet — December 3, 2025
Eric Schmidt argues the real AI disruption will come from automating the “boring” backbone of business—billing, accounting, logistics, and more—unlocking value markets still underestimate. He also points to medicine, climate, and engineering as ripe for breakthroughs, even as skeptics warn of bubble dynamics. The takeaway: the enterprise back office may be AI’s next big profit center.
Three best friends built an AI startup — and made a pact to drink one beer if it all collapses
By Lee Chong Ming — December 3, 2025
Raylu, a startup building AI agents for private‑market investors, raised $8 million in Series A (total $12 million) and plans to nearly double its 11‑person team. The founders say their edge is trust and speed—plus a “one beer” ritual to keep friendship above the business. Their founder lessons: find true product‑market fit, prioritize people, and be willing to pivot hard.
KI‑Update: ChatGPT‑Werbung, KI‑Persönlichkeit, Chatbot Ello, Apple‑Video‑KI
By Marko Pauli, The Decoder — December 3, 2025
The update notes code hints that OpenAI is preparing ads for ChatGPT but may delay rollout while it prioritizes next‑gen models, and warns Europe’s grids could strain under AI’s soaring power needs. Amazon unveiled its Trainium3 chip promising better performance per watt to challenge Nvidia. It also highlights an Anthropic “Soul Doc” leak shaping Claude’s personality, Mistral’s new open models, and Apple’s alternative approach to video generation.
Europa sabe que tiene perdida la carrera por los grandes modelos de IA. Así que Mistral lo está apostando todo a lo pequeño
By Antonio Vallejo — December 4, 2025
Mistral launched its Mistral 3 family: a flagship Large model and a suite of compact “Ministral” models under Apache 2.0, designed to run locally on modest hardware. The thesis is “distributed intelligence”—small, tuned models that are cheaper, privacy‑preserving, and good enough for most jobs, from factory robots to offline drones. It’s a pragmatic European play: prioritize ubiquity and flexibility over sheer scale while building a services ecosystem around agents, code, and deployment.
New study shows how AI could unlock deeper heart data from the Apple Watch’s optical sensor
By Marcus Mendes — December 2, 2025
Apple researchers propose a hybrid modeling approach that infers cardiovascular biomarkers—like stroke volume and cardiac output—from PPG signals by mapping them to simulated arterial pressure waveforms. In tests, the method tracked trends accurately (though not absolute values), outperforming conventional techniques. It hints at future, passive, long‑term heart monitoring using sensors the Apple Watch already has.
Dan Ives says AI bubble fears are overblown and he’s betting on this one data center stock
By Chris MacDonald — December 3, 2024
Analyst Dan Ives argues AI infrastructure spending will accelerate into 2026, supporting bullish cases for data center operators like Iren despite premium valuations. He notes strong Wall Street targets and a still‑intact growth backdrop, suggesting “AI bubble” concerns are overstated. The risk: rich multiples; the bet: rising monetization as deployments scale.
Tim Cook reportedly lost confidence in Apple’s AI chief — can an ex‑Microsoft CVP push Apple Intelligence further?
By Kevin Okemwa — December 3, 2025
Apple hired Amar Subramanya, formerly a Microsoft AI CVP and long‑time Google engineering leader, as VP of AI amid leadership changes following John Giannandrea’s retirement. Reporting to Craig Federighi, he’ll oversee foundation models, ML research, and AI safety and evaluation. The move signals Apple’s intent to push beyond incremental features toward more ambitious Apple Intelligence capabilities.
