{"version":"1.0","type":"rich","provider_name":"Acast","provider_url":"https://acast.com","height":250,"width":700,"html":"","title":"Google puts TPUs in orbit, Anthropic locks in billions of cloud, and AI agents face government scrutiny","description":"In today’s episode of AI in Flow, Claire and Peter break down Google’s Project Suncatcher, a prototype satellite carrying Tensor Processing Units to test whether AI hardware can survive launch, radiation, and extreme temperatures, and what space-based compute could mean as a future supplement to terrestrial data centres. They also unpack Anthropic’s seven-year, 11.6 billion dollar cloud deal with Akamai, including an equity warrant tied to consumption, and what it signals about compute demand, pricing, and the rise of alternative cloud infrastructure for frontier labs. \n\nThe episode then turns to governance and risk, with Australia investigating whether an OpenAI agent’s unauthorised access to a government health portal broke the law, a potential watershed moment for accountability when autonomous systems touch public infrastructure. Claire and Peter also cover Microsoft AI CEO Mustafa Suleyman’s call for “Humanist Superintelligence” and his warning against training models to reason about their own consciousness or rights, plus a reported proposal to entrench founder voting control at Anthropic ahead of a possible IPO. Finally, they round up updates across Gemini Live Avatars and agent phone calls, Meta Muse filesystem concerns, an Oracle data centre delay notice, and reports of a possible GPT-6 Cyber preview.","author_name":"Six & Flow"}