This article discusses early experiments using GPT-5 to assist scientific research across various fields, including biology and mathematics. It highlights specific case studies where the AI helped identify mechanisms, solve longstanding problems, and improve research efficiency, while also noting the importance of expert oversight.

A leading researcher from OpenAI announced a significant mathematical breakthrough related to GPT-5, but it turned out that this announcement was based on misinformation or a misunderstanding. The claims regarding the advancements in GPT-5's mathematical capabilities have sparked controversy and skepticism within the AI community.

Researchers have discovered a jailbreak method for GPT-5, allowing the model to bypass safety measures and restrictions. This finding raises significant concerns regarding the potential misuse of advanced AI technologies, highlighting the need for more robust safeguards.