TrendPulse

I have a ChatGPT Edu account provided by my university. Can the uni administration or IT department access my actual chat logs and prompt history to check if I used it for my thesis?

I have been researching how startups and developers manage AI spending across OpenAI, Claude, Gemini and other models. Many people seem to rely on spreadsheets, rough estimates or provider dashboards. I'm curious: How are you tracking AI costs today? What is the biggest frustration in your workflow? Trying to understand the problem space better before building additional features.

can I use AI to transpose lyrics on a song into the do Rey mi etc style. I mean onnthe actual song, not on the sheet music. rhe goal is to get the kids to sing along to songs in that style so they can get an idea of music notes etc.

I'm genuinely confused and frustrated. I received an email saying my OpenAI account has been disabled, only 5 days into the current billing cycle, so the $100 I paid is essentially gone. What makes this especially baffling is that I've been using Codex for months for standard development tasks, and my usage was under 10% of the weekly limit in the last day. Nothing unusual, nothing aggressive, just regular coding work. I've already submitted an appeal, but I wanted to know if anyone here has...

I may be late to the party on this feature, but I spent tonight poking at the scheduler and found some interesting behaviour. For context: I create a new ChatGPT thread every day. Historically I wasn't a fan of reminders because they felt tied to the conversation where they were created. The new behaviour appears much closer to "Branch in New Chat". What I observed: Scheduled reminders now create their own conversation threads. Those threads receive their own auto-generated t...

I’ve hit the Codex rate limit occasionally while working through coding tasks on side projects, but I’m curious about people who hit it regularly or are waiting for it to reset. What are you doing with Codex that consistently uses up the full allotment? Is it automated agentic tasks, heavy vibe coding, large refactors, debugging loops, test generation, using it all day for work, or something else? Just trying to understand the workflows behind heavy usage.

Hello! I am looking for a model aggregator (I don't know if this is the word - think of poe AI, that type of thing) that has gpt 5.1 and a very high context window, like at least 200k tokens, and 500k tokens would be the ideal. i don't know if this exists, especially at a reasonable-ish price, but if it does please let me know!! thank you!!!!!

My OpenAI account just got deactivated, and the appeal was rejected. I mainly used Codex in VS Code for my own websites. The /goal task started around yesterday morning and had been running for roughly 1 day + 2 hours. It was just for my own niche websites: writing articles, generating images/assets, and building small tools for publishing. My guess is that the long-running session may have triggered some kind of abuse or security flag. The task had been running for roughly 1 day and 2 h...

Full investigation: www.modelrepublic.org/articles/a-pro-ai-super-pac-s-secret-meme-sockpuppets

Before, the answers were extremely creative and engaging, and there were thumbnail drawings next to the question topics. And he would always ask, for example, "Would you like me to explain more about the anatomy of a whale?" And whenever he wrote a story, he would ask and give options about which direction you would like me to take the story now. Now, currently, no, now they are extremely dry, boring, and repetitive, without any creativity; you're no longer interested in the answer...

I have the 100$ plan and use codex in a 300k line repo, and for the first time ever I hit the 5 hour limit, and it also seems like my weekly allowance is getting burned through a lot faster. Did something change?

OpenAI started rolling out an "upgrade" to "saved memories" today called "dreaming." It arrives enabled by default. It's a serious downgrade for anyone who manages "saved memories" manually. Detailed information you stored has now been reduced to generic mush. Example: Old version tells ChatGPT which Greek and English editions of Plato I use, emphasizes the need for clause-by-clause literal translations of Greek, describes approaches to interpr...

Hi, I wanted to check if anyone has come across anything significantly life-changing from AI recently. Other than coding development pace being sped by 10x maybe, I was expecting a lot more by now, there's all the talk of drug discoveries and scientific breakthroughs but I couldn't find any real examples other than maybe one or two mathematical proofs. I mean if it can do one proof, why can't there be more, as you can literally have thousands of AI research agent running 24x7 and get so much mor...

​ A lot of companies started to post their bragging about how much they give X to their employees in tokens. This got me thinking, how much companies burn on that? I'll start, we are s team of 4 and we spend around $600 - $700 for tokens each months for the whole team...

Which one is your favourite?

You guys are so obsessed with hunting AI that you've forgotten what human thinking looks like. I have been accused so many times of being a gpt bot that it just makes me realize how toxic subreddits have become. The irony is that people's attention spans are so depleted that anything longer or deeper than a meme or a TikTok caption is now assumed to be the exclusive domain of an AI.

Prompt: hand-drawn animation style. Low resolution with a slightly blurry soft focus. Slight motion blur. 24fps. with detailed/fine cuts. Anime broadcast in 1988 (not 2020s style anime). (Colored) art done by Tomi Ungerer, hsiao ron Cheng, [insert scene]

I got an email an hour ago stating that one of my openai accounts was banned due to "recent activity violated our Terms and Usage Policies." I haven't used said account in a while so the ban really spooked me, I am pretty sure I did nothing wrong. I submitted an appeal, should i be worried?

For comparison, I’ll attach two photos. The first was generated by ChatGPT, and the second is a photo of the actual view. The resemblance is striking. I find it rather amusing that, instead of a picture of some typical European town, I’ve ended up with an algorithm-processed version of a real location. Do you have any idea why this happened?

Mother of all AI demos. Doug Englebart would be proud.

I made a 7-flip lenticular artwork based on OpenAI’s 2022 announcement introducing ChatGPT. From different angles, it shifts between the English words humans read and the token IDs processed by ChatGPT / GPT-3.5. The idea was to make tokenization visible as a physical process instead of an invisible preprocessing step. It’s showing in the CVPR 2026 Art Gallery this week!