GOOGLE IS SHIPPING GEMINI MODELS FASTER THAN ITS AI SAFETY REPORTS

Summary from the TLDR AI newsletter

Google has launched an AI reasoning model, Gemini 2.5 Pro. The model leads in coding and math capabilities, but Google hasn't released safety reports yet. Google plans to publish these reports after collecting feedback from experimental releases, though this approach raises transparency concerns. Despite pledges for transparency, Google seems to prioritize speed in model deployments, diverging from industry norms for responsible AI practices.