I put GPT-5.2 through an 13-round test, and the AI model raised some serious questions (www.zdnet.com)

A recent evaluation of OpenAI's GPT-5.2 model revealed significant insights through a 13-round testing process, where it faced a series of challenges related to text comprehension, academic explanation, math, and creative tasks, among others. While GPT-5.2 showcased strengths in explaining complex concepts for younger audiences and generating original narratives, it was marked down for some deficiencies, notably in coding accuracy and unnecessary prompt confirmations. The model demonstrated inconsistencies in performance, such as scoring lower in a coding task despite being positioned as an improved version for such functionalities. This assessment is significant for the AI/ML community as it raises questions about the evolution of generative models and their practical applications. The findings suggest that while there are positive strides in narrative creativity and conceptual analysis, essential functionalities like coding need improvement. The need for repeated user confirmations for responses, coupled with performance fluctuations, indicates that GPT-5.2 may represent more of an incremental upgrade rather than a groundbreaking advancement over its predecessor, GPT-5.1, challenging the perception of AI's rapidly improving capabilities.