OpenAI is reportedly testing GPT-5, indicated by recent leaks and mentions in their internal BioSec Benchmark repository, suggesting its application in sensitive areas like biosecurity. While there’s no confirmed release date, speculation suggests an official rollout in the coming months. Features of GPT-5 promise to unify memory, reasoning, vision, and task execution, allowing users to perform multiple functions, such as scheduling and planning, from a single prompt. Enhanced safety measures are anticipated, particularly with its involvement in biosecurity research, which raises concerns about misuse. ChatGPT Agents have intensified curiosity about GPT-5, as they bring substantial capabilities like online scheduling and booking. Users hope that GPT-5 will resolve issues related to hallucinations and poor comprehension observed in previous models. OpenAI is expected to offer GPT-5 first to subscribers in higher-tier plans, ensuring it meets quality standards upon launch.
Source link
