Openai's Astra model eliminates human spatial reasoning edge

OpenAI's latest model, Astra, has reportedly closed the gap in spatial reasoning, a previously recognized human advantage. Boris Power, head of applied research at OpenAI, stated that this cognitive edge no longer exists. Independent researcher @spicylemonade confirmed Astra's flawless performance in spatial reasoning tests, marking a significant advancement over previous large language models. Astra, the latest flagship product from OpenAI, was launched after about a year since the release of the GPT-5 series and two months since the GPT-5.6 update. OpenAI described it as a historic leap in professional work and called it 'the smartest and best-aligned model in the world.' The model's capabilities extend to automating a wide range of tasks, from filling out tables to scheduling appointments, demonstrating its potential for all-around automation. OpenAI's president Brockman hailed the launch as the beginning of the AGI era, emphasizing the new heights of human-machine collaboration. The model's release has sparked discussions about the future of AI and its implications for various industries.