How to analyze a conversation with ChatGPT (GPT-5) to know which answers are based on history and which ones are just suggestions?

which of the model’s answers actually use the real history of my previous conversations

which answers are just general suggestions,

and which ones might be unfounded extrapolations.

It’s really important for me to get a reliable analysis without any made-up information. I’m looking for:

• a concrete method to analyze an AI conversation,
• tools or a process to distinguish “the model is truly using my chat history” vs. “the model is inventing or making overly broad deductions,”
• and ideally, the opinion of an AI/NLP/LLM expert who can explain how to verify this properly.

Additional context:

• The conversation is in French.
• It contains several questions and answers.
• I want to avoid any wrong or inaccurate interpretation.
• I can share an excerpt or even the entire conversation if needed
• My question is how can you reliably analyze a conversation with an LLM to determine which answers genuinely come from history and which ones are just general suggestions?

Thanks in advance for any help, methods, or expertise.