OpenAI has released GPT-6 Astra, a significant advancement in artificial intelligence. President Greg Brockman believes it represents a generational leap and potentially the arrival of artificial general intelligence, or AGI. Astra's capabilities allow AI agents to perform complex professional work autonomously. The model was trained on an unprecedented scale, utilizing over 100,000 GPUs. Notably, Astra's training incorporated other AI models for supervision. It will be available to select organizations and then more broadly to ChatGPT users and developers. Astra is the first model to meet OpenAI's critical cybersecurity threshold, capable of finding vulnerabilities without human guidance. The model can integrate directly into software to perform tasks like formatting legal documents and building games. Demonstrations showed Astra creating 3D scenes, designing circuit boards, and assisting with scientific discoveries. Real-world effectiveness and safety concerns remain to be fully assessed. Concerns persist about the rapid pace of AI development outstripping societal adaptation. OpenAI acknowledges Astra is harder to monitor, making it a research priority to improve oversight mechanisms.