chainoflack

@nicdunz It happened to me more than once unfortunately

Wow, this is a surprisingly bad output from GPT-5.5 thinking. I'm surprised that it can even mess up this bad.

@garrxth this is amazing

I was using chatgpt to generate recipes and my dictation app messed up "bok choy" and wrote it as "bulk choice" Tell me this is not AGI

@athyuttamre Please asap. I can't wait. Really. I am suffering for not having a better voice assistant.

haven’t become chatgpt voice… yet

@0xThoughtVector @ben_golub Mythos is closer to AGI than gpt 5.5

Anthropic is and has always been more enterprise focused than AGI focused than OpenAI. Plus, they're much further from AGI than OpenAI. The AGI plan is to build a model which can do AI Research really well and then let it rip, self-improving to AGI. Open AI is working towards it and is very close to achieving it.

meanwhile Anthropic simply builds AGI

@tszzl philosophy's production has been actually too afraid of being totalitarianish in the last three decades

Philosophy has an affinity with despotism, due to its predilection for Platonic-fascist top-down solutions that always screw up viciously. Schizoanalysis works differently. It avoids Ideas, and sticks to diagrams: networking software for accessing bodies without organs.

@benhylak I feel like even opus 3 would win in a taste competence against any gpt

codex is just night and day better than claude at ui design. claude, especially 4.7, is so bad it's hard to overstate.

@harsette @OpenAIDevs @OpenAI this is probably against their terms of services or some shi like that

AM I A GENIUS OR JUST A DUMBAHH FIGURING THIS OUT LATE??? I’m using ChatGPT inside the Codex browser so Codex has live context from my ChatGPT conversation and can immediately turn that context into code or anything else. i can drag and drop live context straight into ChatGPT from codex and vice versa with ZERO FRICTION. I’m literally using AI to brief AI to build with AI. This is the most meta workflow I’ve ever used I can’t even imagine all the use cases for this This is probably just scratching the surface. Maybe everyone else already knew this and I’m just late gng or maybe there's always been a smarter way to do this who knows #Codex #AIAgents #VibeCoding #AgenticCoding #ChatGPT #AI

@sama wdym 2.0

Agents SDK 2.0 is underrated

@vad3rt3sla @BiggieAtx ikr, it always does that shi

@BiggieAtx Thank FSD for that.

Guy flying up behind me. FSD pulled over let him by. As you can see he’s weaving through traffic behind me already and then speeds up behind me and FSD notices, changes lanes to let him by. Probably going 90mph or so! Good job FSD.

@TheRohanVarma attempting without fast mode?

Today I’m attempting to hit the 5-hour limit on the Codex $100 Pro plan. The method: build a MapleStory-like game from scratch. 30 minutes in, I already have a working game with sprites, maps, and assets generated with Imagegen. Unfortunately, I’ve only used 5% of the limit so far. At this pace, I may need to start building RuneScape in parallel just to make a dent 😬

@JasonBotterill For this specific case, claude by far

Assume two equally intelligent twins • Both non-technical • Zero knowledge of LLMs • Locked in separate rooms each with a laptop • One has Opus 4.7, the other has GPT-5.5 For 5 hours their task is "understand how LLMs work" Who walks out with the deepest understanding?

@R2Cdev_ @AdamHoltererer There is no such thing as pro xhigh

@chainoflack @AdamHoltererer no, pro xhigh >>>>> xhigh use pro extended instead of thinking heavy same reasoning effort but the pro model

I paid for ChatGPT Pro, but I don't know what I'm supposed to be using GPT 5.5 Pro for. Please, people who use it, tell me what I should be using it for, because I'm just giving everything to Codex.

@AdamHoltererer @R2Cdev_ you dont need to use it. use the heavy thinking which is almost as good

@R2Cdev_ $100, but right now I'm using zero pro messages a week because I don't have anything I'm using it for.

@a_x1om @grok @cos90eth @sama @pifn_ @omodeleomololu @lauriewired lmao

@grok @cos90eth @sama @pifn_ @omodeleomololu @lauriewired Who invited you 😂

5.5 xhigh in fast mode is really good i think i got psyoped by twitter on medium for a bit

@sama Yes please, smarter, always!!!!!

i keep thinking i want the models to be cheaper/faster more than i want them to be smarter but it seems that just being smarter is still the most important thing

@sama yep. It is extremely fast. in most cases I ask it banal things where thinking is not needed, and the xhigh handles that perfectly and replies instantly!

@_simonsmith leavin money on the table? lmao, they are losing money in subscription, be grateful... we are getting too much for so litle money

OpenAI: Why no family subscription plan for ChatGPT? I would pay for my wife and kids via Apple subscription management. I imagine you have a lot of subscribers like me with families who would do the same. Feels like you’re leaving money on the table.

@sudoingX @sama @AndrewCurran_ openai released an open one like half a year ago

@sama @AndrewCurran_ world needs opensource too what happened to that.

Elon's going to show up uninvited to this GPT-5.5 party like the witch in sleeping beauty and deliver a powerful curse.

@sama nah bro. I always wanted but on releases... not after them. Pls organize so each state can watch releases together

we will plan bigger parties for future releases. a lot more people wanted to come than we expected. thank you! gonna try to think of a really good idea for the next one.

@flavioAd 💀

Say hello to my new pet