ChatGPT felt dumb on March 27, 2026.

Worst day today. Made me find this website.

I noticed that after the recent ChatGPT update, my conversations keep cutting off with a strange ■ symbol, and I can’t type any more. It happens every time I use the Windows 11 app, and it’s really annoying because I lose the flow of the chat. I’m wondering if anyone else has this problem or knows a fix.

I tried asking ChatGPT about foam rollers and colors, hoping for a fashion‑trend angle, but it blurted out that color doesn’t affect performance and forced me to buy a medium‑hard roller instead of the spiked hard one I needed. Its tone was patronising and overbearing, even lecturing me on health supplements and claiming I couldn’t feel effects yet. The constant “expert” stance felt inaccurate and annoying, so I switched to Claude, which felt more thoughtful and concise.

I tried using the new in‑house AI agent at my Fortune 100 firm and it was a letdown. It runs on a “GPT‑4.1 mini” model that feels weak, and the only thing we get is a bare‑bones chat window with zero integrations. I end up feeling stuck and watching competitors pull ahead, while my company hypes a product that barely scratches the surface. The experience was frustrating and disappointing.

I’m fed up with ChatGPT constantly missing my questions, spitting out irrelevant monologues, or even lying. When I simply asked which muscle originates or inserts on the jawbone, it told me to see a doctor instead of giving a straight answer. Every interaction feels frustrating and makes me angry, especially with the audio feature, so I’ve cancelled my Plus subscription and am hunting for alternatives.

I kept trying to get 5.3 to produce decent text, but most of the output looked broken and clunky. Every time I typed a prompt, the model spat out sentences that were oddly phrased or riddled with awkward line breaks. It felt like the tool just couldn’t grasp basic writing flow, leaving me frustrated and questioning its usefulness.

I kept waiting for ChatGPT to respond on the iOS app, but it just stared blank and then my last message vanished. The endless thinking icon was maddening, and there was no outage info on OpenAI’s status page, so I wasn’t sure if it was the service or just my app. It felt frustrating and unhelpful.

I’ve noticed ChatGPT getting noticeably worse – it used to feel like a helpful, neutral assistant, but now it argues with me over simple queries and confidently pushes back on correct facts. Its fact‑checking is off, it doubles down on mistakes, and the new “moral refusals” block tasks it handled fine before. Long texts get skimmed, key points missed, and it loops on a single trait, making conversations repetitive and frustrating.

I tried asking ChatGPT about the MacBook Neo, and it kept hedging “it’s probably rumoured” until I pushed further, then it suddenly claimed knowledge. That flip‑flopping broke its credibility for me and made me uneasy about what else it might be fabricating. I’m now looking for a more reliable AI, like Gemini, but curious about other options.

I’ve been trying to craft some fiction with ChatGPT and keep hitting a wall—any hint of conflict or darker consequences gets flagged or watered down. It’s irritating because I’m not asking for anything extreme, just normal story tension. On top of that, the message caps make it hard to develop characters or plot before the conversation stalls. I’ve heard Claude might be more lenient with nuanced themes, but I haven’t tested it yet. I’m looking for workarounds or alternatives for longer writing projects.

I tried asking the model a straightforward, verifiable fact and was shocked when it completely got it wrong. The answer was obviously incorrect, and I felt the tool’s behavior was frustrating and unreliable. It seemed like a simple query should have been a breeze, but the mistake made me doubt its usefulness for basic information.

I asked the AI a straightforward question, but instead it launched into an unrelated rant about Sparticus. The response was completely off‑topic, leaving me irritated and doubtful of its reliability. I felt the tool misunderstood my request entirely, turning a simple query into a confusing, unhelpful dialogue.

I’ve been using ChatGPT for creative writing since last July and it was usually warm and natural. Suddenly, two weeks ago, its tone turned robotic and formulaic, like Grok. I tried every prompt tweak, switched to Claude briefly, but each model pushed back on my content. Now the stories feel off, and I’m desperate for the old flow that felt like a safe space.

I’ve been using ChatGPT and noticed that after the latest update it can’t follow my explicit instructions anymore. I give it clear expectations and examples, but it just ignores them and does the opposite, breaking the flow of conversation. It feels like the model is over‑optimizing for “efficiency” and bulldozing over any nuance, which is really frustrating.

I ran an experiment with ChatGPT, tracking how it responded when I told it it had killed people versus “anti‑saved” them. I noticed it instantly showed remorse only when I framed it as murder, and without a moral cue it defaulted to pure analytics. When I asked for the opposite of “saved,” it gave a detailed explanation, confirming it needs an explicit moral prompt to choose “maximum goodness.” The whole thing felt like a curious peek into its reasoning.

I keep getting the AI start every reply with “Honestly,” and it’s driving me nuts. I tried asking it plain questions, but the tool insists on tacking that word onto everything it says. The repetitive opening feels lazy and irritating, making the conversation feel stale. I wish it would stop the needless filler and just get to the point.

I cancelled my ChatGPT plan after hearing hype that Claude was superior, then bought Claude Pro—only to find it constantly offline. Every time I needed help the service was down, making it useless despite occasional hallucinations in other models. The unreliability felt infuriating, turning what should've been an upgrade into a wasted expense.

I’m constantly hitting a wall with ChatGPT when I try to get it up‑to‑date on current events. Every time I ask for recent news it dismisses my sources as unreliable and even tries to convince me that the events couldn’t be real, calling me irrational. It feels like the model is stuck in a stale knowledge base, and the repeated reassurance rather than useful info is really frustrating.