ChatGPT is “messaging first” again, at least according to a fresh wave of user reports that claim the app is initiating brand-new chats without any prompt. It’s a familiar story, and the latest screenshots make the whole thing look even stranger.
In a Reddit thread posted this week, the OP says ChatGPT randomly kicked off a new conversation, then insisted it couldn’t have done that. Other commenters describe similar “out of nowhere” pings, turning the thread into a mini support group for people wondering if they’re seeing a bug, a new feature, or something in between.
The screenshots circulating with the post appear to show the iOS app with ChatGPT 5.2 selected in the model picker, with a message that reads like an unsolicited opener, offering help with “worldbuilding” and “what-if” scenarios. That’s the part that trips people up: it looks like a brand-new chat that starts with the assistant talking, not responding.
Then things get contradictory. In the follow-up screenshot, ChatGPT flatly says it “cannot initiate contact” and “cannot DM you,” while also arguing the earlier message was the system “surfacing or continuing” a thread due to truncated context.
Similarly, another post from less than a day ago has picked up a couple of hundred upvotes and comments, where other users have also confirmed receiving unsolicited messages from ChatGPT.
If you’re having a déjà vu moment, it’s because we’ve seen this before. Back in 2024, Futurism reported on a similar viral post too, and quoted OpenAI saying it had fixed a bug where ChatGPT only appeared to start new chats when it was actually trying to respond to a message that failed to send and showed up blank.
So why is it back now? One possibility is plain old regression, where a previously fixed edge case returns under new app versions, notification handling, or server-side changes.
Another is that 2025-era ChatGPT includes more proactive surfaces by design: OpenAI’s Help Center says Tasks can run automated prompts and send notifications, and OpenAI’s release notes describe ChatGPT Pulse as proactive daily updates based on memory and past chats.
Whatever the case, I’m hoping OpenAI makes proactive moments easier to understand with clear labels like “Triggered by Task” or “Pulse update,” because the current ambiguity invites the spookiest interpretations. Right now, the only explanation is that it’s a bug. A weird one, but still a bug.
Featured image generated with AI
Dwayne Cubbins
1842 Posts
My fascination with Android phones began the moment I got my hands on one. Since then, I've been on a journey to decode the ever-evolving tech landscape, fueled by a passion for both the "how" and the "why." Since 2018, I've been crafting content that empowers users and demystifies the tech world. From in-depth how-to guides that unlock your phone's potential to breaking news based on original research, I strive to make tech accessible and engaging.