Lessons from Studying Every Memory System — Shlok Khemani, Independent
Early ChatGPT memory forced users to manage memory creation, burdening rather than helping them.
“as a user, because you could see every time a memory was created, it felt like you were responsible for both creating memories while you were just trying to have a conversation”
Independent researcher Shlok Khemani, who spent a year reverse-engineering memory systems in ChatGPT, Claude, and Gemini, presents a historical analysis of how AI memory evolved from stateless chat to persistent personalization. His key critique of ChatGPT Memory v1 (Feb 2024) is that making memory creation visible to users inadvertently shifted the cognitive burden of memory management onto them. This is a practitioner-level engineering retrospective relevant to teams building personalized AI products.