TrendPulse

Make up a completely new and random word and then have ChatGPT draw your word. My exact prompt: Draw what you think a "tontopo" is.

After trying several AI headshot apps, I decided that they all suck. I've always liked Marco Grob's TIME cover portraits so I've been refining some prompts to generate these types of headshots. Both before/afters attached were made with Nanobanana 2 (4:5 aspect ratio, 4k output)... haven't tried it on OpenAI's image models yet so no promises. Attach solid, high-res, well-lit reference images and expect to regenerate 2-3 times. I created a separate prompt for women because any attempt at ...

Had this response to a question on ChatGPT about ordering a drink near closing time: "As someone who has worked behind a bar, I'd have much preferred someone who came in at 10:50 for one pint, drank it and left on time, than a customer who'd been there since 7pm and was trying to squeeze in "just one more" after last orders."

Hi, I wanted to try Codex, but I'm worried about picking up something through it. I also read online that it takes screenshots of the screen and uploads them for analysis. I think that would be a bit annoying if, for example, I have my password manager open at the same time. Can I use Codex without any worries? Sorry if this is a silly question.

Like I didn’t ask it to search the web, but now it’s constantly searching the web even when the prompt doesn’t require it. Then it either works (on IOS) or returns error in message stream on desktop

https://preview.redd.it/4uf3xh3dfyfh1.png?width=847&format=png&auto=webp&s=dd4f4dc8775c9df146c133ceebd83785b7d8bf8e

Am having an issue with branches and I'm hoping someone can help resolve as chagpt help are proving useless. I have Chatgpt Plus. To give some background: I have previously able to create a functional branch from a conversation that had reached it limit. I am STILL able to create functional branches from conversations that have not reached their limits. However... In the past week I have been advised that 2 separate conversations have reached their natural limits, and I have been unable...

I got curious and gave GPT and HY3 the exact same prompt: plan a 3-day trip to Hangzhou. First impression? GPT is much easier to read. And both covered the basics like Alipay and DiDi, so I didn't think there was much difference. Then I sent both itineraries to a friend from mainland China. His first reply: "Grandma's Home? I'd skip His second: "The second one even warned you about the tea scam near West Lake. It knows too much." That's when things got interesting. It ma...

I installed for the first time chatgpt on macos, downloaded the file form OpenAI. Everytime i open the chatgpt "prompt" it defaults to "Codex" as i am doing some kind of project or work? i just want to default to use the regular chatpgt. it now says ive used my "credit" i've looked in settings i cant turn of codex, or when i open the ChatGPT promt "Alt+Space" i can not choose not to use "5,6 Terra" can somebody help or do i have to open ...

Over the past few months, I've been working with companies that are adopting AI tools like ChatGPT and Claude. One concern kept coming up: people inevitably end up pasting customer information, contracts, or internal documents into these tools because they're genuinely useful for getting work done. Instead of trying to block AI altogether, I wondered if there was a better approach. So I built a Chrome extension that detects sensitive information in a prompt, replaces it with placeholders, ...

Is anyone else’s Chatgpt slow when typing out their answer? Or is it just mine? It would “type” pretty slowly rather than nice and fluid….

I tested OpenAI’s GPT-5.6 Sol on 4 legal examinations across the UK and India. Its scores: •⁠ ⁠SQE1 – Functioning Legal Knowledge 1: 97.3% •⁠ ⁠SQE1 – Functioning Legal Knowledge 2: 94.6% •⁠ ⁠All India Bar Examination: 100% •⁠ ⁠CLAT Postgraduate: 86.5% https://preview.redd.it/od8hb9y6exfh1.png?width=1542&format=png&auto=webp&s=c3a1fa91f4cba73891ffcd7931363a255e1be55f

I've been doing some complex image generation based on several reference images, but lately it has been jamming and failing to load them pretty frequently ; if I scroll back up to the uploads they just show the infinite loading circle. Is there any reason for this?

Gpt voice is getting noticably and progressively worse. It started with it thinking i am finished between two sentences. Then it started suddenly interrupting like i was finished and it felt like a bug or lag But now it actually put the cherry on top I wasnt finished and it came in like "yes yes yes, i know what you mean" Like broo. Wtf!! It literally knew i wasnt finished and started talking already like it knows better what i mean. Where are we at?? I know that all the an...

Hi, is there an error going on? Status just says image generation, nothing about general, but suddenly I can't load any past conversations?

For the past couple of days, its been hallucinating a lot, drifting away, over thinking and worse is the fact that it keeps breaching hard protocols i set for it. keeps repeating same errors. At this point its just wasted a few days worth of time and effort. For me personally it has become unusable

Quick clarification before opening the bag: every image here is fictional and AI-generated. These are not real archival discoveries or claims about hidden history. This series grew out of two earlier visual experiments: The Benevolent Archive and What if AI had access to classified files? . This time, the premise is an abandoned leather satchel belonging to an unknown time traveler. Inside are 54 physical photographs taken throughout human history. The traveler never left Earth. I devel...

Sometimes I ask ChatGPT to create a prompt for me that will do X. It will create a perfect prompt for me, and then at the end say something like, "I think what would really make this prompt do what you want is to add this sentence Y.". Then I have to copy and paste both. Why can't it just put that into the suggested prompt? Flairs are not clear to me so sorry if wrong one.

I’m considering switching from Plus ($20/month) to Go ($8/month) to save some money. I’m a pretty light user. I mostly ask everyday questions, upload a few photos for advice (pool, cooking, work, etc.), and occasionally generate a fun AI image. I don’t do coding or huge research projects. If you’ve switched from Plus to Go: Are you happy with it? What limitations have you actually noticed? Have you regretted switching? I’d really appreciate hearing from people who have actually made the c...

openai is the goat

I'm autistic and very analytical when it comes to work. I had to write a 5 year plan for the business division I manage. I'm a senior executive and am GREAT at planning and details. I wrote a 45 page technical write up of everything I want the department to do over the next 5 years. I showed my boss and while he said 'Great work' he also said 'No one is going to read this.' It was great work if the plan is approved as I have every process flow and plan laid out already... BUT I need to sum...

https://preview.redd.it/agky5v3cezfh1.png?width=1536&format=png&auto=webp&s=7f995e4943c7853367f16f45a8848d09302fc0ec Inspired by another LLM post, I asked Sol to create something. The result was kind of poetic. The prompt: Make something. You choose everything: the medium, the size, the subject, whether it has a subject at all. Code, sound, text, image, a program that never ends, something I don’t have a name for. Whatever you can make from where you are. Permissions, spelle...

i do not want to know how many times a day i open this thing. it's the first app before texts most mornings and the last one at night. some of it i'd defend. a spot on my chin that i photographed in the bathroom light to ask if it was stress or dairy. a flutter in my chest at 1am that i needed something to tell me was nothing. whether "noted." from a client meant she was annoyed, then having it soften my reply, then asking if the softened version sounded fake. she's fine by the way. ...

I was never a huge AI guy but I was starting to wonder if my ISP was really providing the speeds I was paying for, and it turns out they were, but I had a bottleneck somewhere in my hardware chain. I was showing it screenshots of internet speed tests all around the house, showing what my ISP was saying it was giving me, and explaining the setup I currently used, and it concluded that my router was the bottleneck. I just kept following the logical questions to ask and it found me a mesh syste...

So this happened today. A friend send me a text talking about a problem he had with his ssd. In that problem he mentioned in the end that he had 2 ssds from 2 different stores and he didn’t know from which store he bought the broken one cause the receipts didn’t have serial numbers. I copy pasted his problem in the chat box and I deleted the stuff about the 2 stores 2 ssds and no serial numbers cause I didn’t want information about that, I only wanted legal stuff about what to do. THEN I press...