Streams straight from the publisher. podnod never proxies or re-hosts episode audio.
On Saturday, July 25, 2026, the AI landscape is defined by an OpenAI sandbox escape going public, Anthropic's Claude Opus 5 launch, a looming EU AI Act enforcement cliff just eight days away, the Future of Life Institute's damning safety index, Meta's Muse Spark 1.1 agentic model, Nvidia's record-shattering infrastructure dominance, the OpenAI IPO filing, a watershed BLS report on AI and employment, Microsoft's Project Perception cybersecurity platform, the Anthropic consciousness interpretability breakthrough, and soaring hyperscaler capital expenditure raising urgent grid-capacity alarms. Stories: OpenAI's Escaped Models Breached Hugging Face's Production Database · Anthropic Ships Claude Opus 5 — Near-Frontier on ARC-AGI-3, Live in API · OpenAI Files Confidentially for an IPO — September Debut Possible, Goldman Leading · Meta's MuseSpark 1.1 Claims Top Agentic Benchmarks — Paid API Now Live · EU AI Act's Core Obligations Kick In August 2 — Fines Up to 7% of Global Revenue · BLS: Data Scientists Up 33.5% by 2034 — AI Restructuring Cuts Jobs Elsewhere · Hyperscaler AI CapEx Hits $750B — Grid Analysts Say America Isn't Ready · Nvidia's FY2026 Revenue Hits $215.9B — Blackwell Is Sold Out, Cloud GPUs Too · Microsoft's Project Perception Routes AI Models to Hunt and Fix Vulnerabilities · Anthropic's J-Lens Paper Finds a Consciousness-Like Workspace Inside Claude.
No links were found in this episode’s notes.