{"version":"1.0","type":"rich","provider_name":"Acast","provider_url":"https://acast.com","height":250,"width":700,"html":"<iframe src=\"https://embed.acast.com/$/68a43f4573bf5b62987006aa/6ae96d1c85d3a4476296d6f9?\" frameBorder=\"0\" width=\"700\" height=\"250\"></iframe>","title":"OpenAI tests grid relief, security sprints, and the legal fight over scraping","description":"In today’s episode of AI in Flow, Claire and Peter unpack the latest signals that AI’s next phase is being shaped as much by infrastructure, security, law, and finance as by model releases. OpenAI pilots a Grid Protectors programme at a Georgia data centre, aiming to free up to 1,000 megawatts during emergencies, a potential template for winning community and utility support for new compute campuses. They also cover OpenAI’s expanded security push, including a new sprint with Cathedral, bigger bug bounty payouts, broader model monitoring, and subsidised access for defenders via the Daybreak initiative.\n\nOn the legal front, the New York Times and other publishers sue Brave over alleged scraping and paywall bypassing, raising the stakes for any business relying on scraped indexes or real-time content pipelines. Plus, Microsoft returns to the bond market with a $25 billion sale to fund AI and cloud expansion amid huge capex plans, alongside workforce reshaping. Finally, India launches a new DEPA-based handbook for privacy-preserving AI training, built around the idea that the model travels to the data, not the other way around.","author_name":"Six & Flow"}