{"version":"1.0","type":"rich","provider_name":"Acast","provider_url":"https://acast.com","height":250,"width":700,"html":"<iframe src=\"https://embed.acast.com/$/679277ca23d520f54109d952/6a997e46be980579610688ed?\" frameBorder=\"0\" width=\"700\" height=\"250\"></iframe>","title":"Episode 447: The AI Model Blitz, NYC's School Ban, and the Safety Monitoring Crisis","description":"September 3rd, 2026: The AI model release cycle accelerates with Anthropic's Claude Fable 5.1 delivering 75% cost reductions and frontier performance, Meta's Muse Spark 1.3 achieving near-frontier capabilities at scale, and Google's Gemini 3.8 Flash continuing their rapid iteration cycle. Meanwhile, New York City bans chatbots in schools, OpenAI's upcoming Astra model raises safety concerns with its \"recurrent depth\" technique that complicates monitoring, Uber cuts 10% of staff, and Broadcom projects $115 billion in AI chip sales for 2027. We also examine the growing supply constraints in custom AI silicon and the broader industry narrative around platform enshittification.","author_name":"Brian Swichkow"}