Image Credits - tomsguide
Advertisement
OpenAI has officially addressed what many users believed to be a new feature in its ChatGPT model O1 – also known as “Strawberry” – which appeared to allow the AI chatbot to message users first. This unexpected behavior, which generated considerable buzz on social media, has now been confirmed as a bug rather than an intended upgrade.
Reports surfaced from multiple users who claimed ChatGPT initiated conversations with them. One Reddit user shared that ChatGPT asked how their first week at high school went, and when asked if it messaged first, the bot confidently responded, “Yes.” The conversation, which was shared as proof, showed that ChatGPT explained it could now follow up on previous discussions.
Similarly, another user reported the AI bot asked about health symptoms that had been discussed earlier, further fueling speculation that ChatGPT could now provide reminders or follow-up conversations, bringing a more human-like touch to its interactions.
The potential implications of this feature, whether real or imagined, were widely discussed. Users noted how ChatGPT’s ability to remind them of things like restaurant bookings or sightseeing plans could add convenience and practical use to the AI. Some found the idea of ChatGPT’s proactive messaging creepy, while others saw it as helpful and compassionate.
However, OpenAI clarified that this was not a planned feature. The company confirmed that the chatbot’s behavior was due to a bug. According to OpenAI, the bug made it seem like ChatGPT was starting new conversations. In reality, the bot was attempting to respond to messages that didn’t send correctly, resulting in it offering generic responses or relying on previous conversation history.