What you'll learn

• 1 (00:08) **Second Wave of AI Releases Reveals Key Industry Trends** - The host introduces a batch of new product launches that highlight three major trends: the rise of voice interaction, complex model architectures optimizing for cost, and a shift toward what users will soon be doing with AI.
• 2 (01:39) **Anthropic Report Details AI Misuse and Distillation Attacks** - Anthropic published a report on detecting and countering AI misuse, revealing major distillation attacks from Alibaba, DeepSeek, and Xiaomi, who used fraudulent accounts to extract reasoning traces from Claude models.
• 3 (04:25) **Sam Altman Signals Openness to AI Slowdown** - During an all-hands meeting, Sam Altman told OpenAI staff he is open to pacing the development of new AI models, likely in coordination with other labs, marking a shift from his previous non-committal stance.
• 4 (05:51) **OpenAI Pauses Pro Subscriptions Due to Compute Constraints** - OpenAI product leader Thibaut Socio announced a pause on new subscriptions to the $200 Pro Plan because demand for GPT-6 Astra is "unprecedented" and putting the most strain on their systems.
• 5 (07:23) **Microsoft Plans to Triple Data Center Capacity by 2032** - Sources told Bloomberg that Microsoft aims to have 38 gigawatts of global capacity by 2032, up from 12 GW, with a third dedicated to AI compute and new CPUs for agentic workflows.
• 6 (08:22) **Nvidia CEO Jensen Huang Doubles Down on 70% Growth Forecast** - At a Goldman Sachs conference, Huang said growth is limited only by the supply chain, and that Nvidia's rack-scale systems are growing 27% month over month.
• 7 (09:26) **DOJ Investigates Nvidia's $20 Billion Groq Deal** - The New York Times reports the DOJ has been investigating whether Nvidia's deal with chipmaking startup Groq was structured as a non-exclusive licensing agreement to circumvent antitrust laws.

+ Full timestamped outline available in the app