TechFlow news, according to Cointelegraph, a Reddit user with the username "SentuBill" said that ChatGPT's large language model (LLM) initiated a conversation by asking how his first day at school went. Based on the provided chat log, the LLM was able to infer from prior conversations with the user that it was indeed his first day at school. The large language model also indicated these capabilities were part of a new upgrade, which has been verified as authentic through official ChatGPT conversation records.
Previously, OpenAI rolled out a highly anticipated preview upgrade of its smarter models to select users, stating that these new models would feature reasoning abilities and grant ChatGPT what appears to be a magical "think before answering" capability. Response times for the new models will range between 10–20 seconds, and they will be capable of searching the internet and conducting independent in-depth research.
