I pushed ChatGPT toward its hidden chat limit — here's what happened (www.techradar.com)

A recent exploration into ChatGPT's chat limits revealed that sessions are constrained not just by time or message count, but by a more complex system involving tokens—units of text that can lead to memory restrictions. Tokens, averaging about 0.75 words, dictate how much context ChatGPT can maintain in a conversation, with a context window likely measured in hundreds of thousands of tokens. Users have discovered that, upon reaching this limit, the model starts to forget earlier parts of the conversation, and in some cases, an extended chat can result in an abrupt termination with a prompt to start anew. This revelation is significant for the AI/ML community as it highlights the underlying challenges in managing long-form interactions with AI models. As users increasingly rely on these systems for complex dialogue, understanding the structure of limitations and how to proactively address them becomes vital. Users are encouraged to regularly summarize and recreate conversations before reaching the limits to preserve context and ensure seamless interactions, thus opening discussions about the importance of memory management in future AI designs.