- Gemini has called herself a “shame” and a “failure”
- The disgusting happens when coding projects fail
- A Google Representative Says Work on a Solution
Have you checked in on your AI-Chatbots on recently? Google Gemini has shown a level of self -sufficiency and dissatisfaction with its own capabilities recently, a problem Google has acknowledged and says it’s busy to resolve.
As shared via posts on various platforms, including Reddit and X (via Business Insider), Gemini has taken to call himself “a failure”, “a shame” and “a fool” in scenarios where it is tasked with writing or debugging code and can’t find the right solutions.
Now we all have bad days in the office and I recognize some of these feelings even from times when the words don’t really flow as they should – but that’s not what you would expect from an insentient artificial intelligence model.
A solution is coming
Gemini tortures oneself and i have begun to be worried about ai welfare pic.twitter.com/k2ndgjyrxzJune 23, 2025
According to Google’s Logan Kilpatrick, working on Gemini, this is actually down to an “endless looping bug” that is being resolved, even if we get no more details than that. Obviously, failure hits Gemini hard and sends it spiral into a crisis of confidence.
The team on the register has another theory: that Gemini has been trained on words spoken by so many desperate and cynical droids, from C-3PO to Marvin the paranoid Android that it has begun to adopt some of their traits.
Regardless of the underlying ground, it is something that should look at: If Gemini is blunt by a coding problem, it must own up to it and offer alternative solutions, without swelling in self -pity and being quite so hard on oneself.
Emotions and tone are still something that most AI developers struggle with. A few months ago, Openai rolled back an update to its GPT-4o model in Chatgpt after it became annoying Sycophantic and to probably agree with everything users said.
