OpenAI has officially announced GPT-5, showcasing substantially improved reasoning, planning, and factual accuracy compared to previous models. In benchmark tests, GPT-5 demonstrated human-expert level performance across multiple domains including law, medicine, and engineering. The system introduces a new architecture that better maintains context over long interactions and shows improved capability to catch its own errors. OpenAI is implementing a phased deployment starting with research partners before wider availability. Industry analysts suggest the advance could accelerate AI adoption in professional services while raising new questions about potential economic impacts.