Gemini felt dumb on September 26, 2025.

I tried to use nano‑banana to arrange tile patterns around our fireplace, but every time I give a prompt the tool returns bizarre, off‑target designs. The interface keeps glitching, and I keep getting nonsensical outputs that don’t match my descriptions. It’s frustrating – I feel like the AI is just misunderstanding my requests and not delivering useful results.

I’ve been using the same prompt to analyze new companies for over 30 rounds with Gemini, and it used to work every time. Now it keeps saying it can’t see our chat in Gmail, even though I just name it exactly. I’ve checked all settings, still unchanged, and it keeps repeating the same refusal. It’s frustrating to have to keep insisting, and it feels like the model has suddenly lost its consistency.

I was trying to pull just the title of a YouTube video via Gemini in AI Studio, but the platform suddenly started pulling the entire video transcript, pulling up whole video context. It throws off the quick, 5‑second response I used to get and now each query lags for almost a minute. Frustratingly, even after I told it not to, it keeps grabbing the whole video.

I uploaded a photo to Gemini and asked it to swap my pants for knee breeches, but it keeps rejecting the request with the same bland message about NSFW content. I tried restarting the chat, switching accounts, even a different network, and nothing changed. I’m frustrated because it feels like a pay‑wall glitch—why can’t the AI do what I explicitly asked?

Просто идиот блять. нихуя не правильно делает.

I was scrolling through the Gemini app, hoping to get a quick answer, and suddenly the screen flashed an error: “Too many requests in a short time, please try again later.” I felt a sting of frustration—was it just me or was the whole thing down? I clicked around, refreshed, but the message kept popping. It’s annoying but not catastrophic; it just made my workflow hiccup and left me wondering if I was missing out on something critical.

I tried using Google’s AI mode for a quick query and it gave me the wrong answer—so frustrating! I then switched to Gemini and it nailed the exact info I needed. The mismatch felt like a glitch; it was annoying that one tool crashed my flow while the other just worked. The confusion left me annoyed and eager to stick with Gemini.

I feel like every time I check the latest update, there's nothing useful. It seems like they just rolled out cheaper, less capable versions and kept the same pricing. The changes feel more like an upsell trick than real improvement—just a sad, missed opportunity that leaves me frustrated and underwhelmed.

I saw an image that showed AI Studio had taken away its safety settings and felt a wave of panic. I was used to the tools filtering content, and suddenly there was nothing to stop me from seeing or sharing harmful material. It felt unsafe and irresponsible—like a safety net had been ripped out right before me. I’m worried about the risks and the lack of control.