Aaron Cohen

@sama Narrow gap between the hype from YouTube tutorials and actual practical use of agents. When something sounds too easy on YouTube it simply is not easy.

what problem do you most hope AI will solve in the future? maybe we can help!

@om_patel5 We shouldn't single out Anthropic but Amazon is, admittedly, a better model for customer service for those wishing to scale like this through automation.

THIS GUY PAID $118 FOR CLAUDE MAX AND WHEN SUPPORT IGNORED HIM FOR DAYS HE SERVED A FORMAL LEGAL NOTICE TO ANTHROPIC'S OFFICE paid for max, had issues, and contacted support. got the classic AI bot loop and no human ever responded so he drafted a formal legal notice and served it to anthropic's office himself the funny thing is that he also used claude to draft the legal notice. this is only one of many legal actions against anthropic this month. for example, an EU subscriber already filed a formal complaint under consumer protection law a billion dollar company charging $118-200 a month with zero human support is becoming a pattern their support bot is useless, their response time is days, and now paying customers are (finally) lawyering up at some point anthropic has to decide if they want to be a premium product with premium support or a premium price with startup tier customer service right now it's the second one

@kimmonismus This protest could have been mitigated or prevented with better comms and a slow roll.

Holy: Leaked audio from a Meta all-hands on April 30: Zuckerberg told employees the company is using them to train AI models before mass layoffs hit. His argument? Meta's engineers are smarter than any external workforce, so having them solve coding tasks internally will make Meta's models better, faster than competitors. The layoffs are expected Wednesday at 4 a.m. Train your replacement, then get walked out. That's the deal now.

@Anuragg52704997 @bcherny

@bcherny Claude Code is down. We are getting API Error 500 now.

@sama What about price/EQ? Seems that's rarely discussed publicly anymore by OpenAI and it used to be a huge talking point.

i get some anxiety not using the smartest-available model/settings. but sometimes i dont mind if it's really slow. i wonder if we should focus more on a price/speed tradeoff relative to a price/intelligence tradeoff.

Codex now works directly in Chrome on macOS and Windows. It’s even better at working with apps and sites in Chrome, and now works in parallel across tabs in the background without taking over your browser. To get started, install the Chrome plugin in the Codex app.

@BytCtrl @om_patel5 Sure, but isn't that kinda of all obvious even before? Why else would they degrade a product?

@om_patel5 Explains a lot. Demand exploded, infrastructure couldn't keep up. Hope they stabilize soon!

ANTHROPIC'S CEO JUST ADMITTED WHY CLAUDE HAS BEEN SO BAD LATELY dario amodei at anthropic's developer conference in san francisco said the company saw 80x growth in Q1 on an annualized basis they planned for 10x. they got 80x that's why: > the throttling happened > the usage limits got nerfed (silently ofc) > opus 4.7 felt slower > max 20x customers were getting less than they paid for > support was replaced with bots > they had to rent GPUs from spacex they literally couldn't keep up with how many people started using claude this also explains the spacex GPU deal. they weren't shopping for compute because they wanted to, they were desperate for it because 80x growth broke their infrastructure and it explains every single pricing change from the last 2 months they weren't being greedy, they were drowning in demand they didn't build for doesn't make the user experience any less frustrating but at least now we know the actual reason 80x growth sounds incredible on a slide deck for the users it meant 2 months of degraded service while anthropic scrambled to catch up and the models still feel weak compared to prime opus 4.6 claude PEAKED when opus 4.6 was released. everything since then has been anthropic trying to scale a product that got too popular too fast while the quality fell off

@aravhawk @OpenAIDevs You nailed it. It an't debug worth a damn. I thought it wa supposed to be the smartest model ever (other than Mythos)

@cohencomms @OpenAIDevs Way, way better. It's better than Mythos (probably). Opus can't reliably do much work other than frontend; especially not debugging.

WHAT THE HELL I LOVE YOU @OpenAIDevs 10X RATE LIMITS ON CODEX IS FRICKING CRAZYYYYYYYY

@DiveshSydney @mweinbach Good to know. I was hoping you would say open router or something like that

@cohencomms @mweinbach Possible through CoPilot

Codex goal feature seems cool Looks like you can give Codex a goal and it’ll continue to work, plan, and test until it’s done? I’m just reading the commits here but that’s what I think it is?

@simranrambles X-or

wonder how will Elon rename cursor, considering both xcode and codex are taken

@sama Everything actually lands in the middle which is far more boring

"post-AGI, no one is going to work and the economy is going to collapse" "i am switching to polyphasic sleep because GPT-5.5 in codex is so good that i can't afford to be sleeping for such long stretches and miss out on working"

@AlexFinn @AlexFinn is the vibe king

ChatGPT 5.5 + ChatGPT Image 2 is the sickest combo Spent 4 hours straight developing a design language for my app. Most fun I had with AI in months You guys have to try this out: 1. Get 4 or 5 inspiration images. Just vibes you like 2. Ask GPT 5.5 what the qualities are of those images 3. Describe the app you're working on 4. Ask to generate a mood board of the inspiration design style mixed with the qualities of your app 5. Keep going back and forth with the AI tweaking the mood board and design language until it feels like it matches what you want for your app. Keep dropping in more inspiration images as you go. Like new spices being added to the dish 6. Spend hours doing this. Keep massaging it until the images and mood boards match the style you want 7. When ready, get a final description and put it into an AI design tool to get a final interface Was a blast doing this for hours and got a final feel/design language for my app HIM Truly the best part about AI is that it takes away all the gatekeeping. All the technical skills necessary. All the knowledge moats. All that's left is pure expression of creativity. Carve off a couple hours tonight and just let your creativity flow

Introducing Claude Design by Anthropic Labs: make prototypes, slides, and one-pagers by talking to Claude. Powered by Claude Opus 4.7, our most capable vision model. Available in research preview on the Pro, Max, Team, and Enterprise plans, rolling out throughout the day.

@sama Saw that subtle dig on Claude 😉

I am happy everyone is switching to Codex, but Tibo if you start rate limiting me or making me use worse models...

@sama Thanks for sharing. I wish it was a bit deeper with details but this is a start. Keep the blogs pumping

I wrote this early this morning and I wasn't sure if I would actually publish it, but here it is: blog.samaltman.com/2279512

@OpenAI Can you provide a comparison with Claude and Gemini? It's hard to compare what you get and who to choose

We’re updating our ChatGPT Pro and Plus subscriptions to better support the growing use of Codex. We’re introducing a new $100/month Pro tier. This new tier offers 5x more Codex usage than Plus and is best for longer, high-effort Codex sessions. In ChatGPT, this new Pro tier still offers access to all Pro features, including the exclusive Pro model and unlimited access to Instant and Thinking models. To celebrate the launch, we’re increasing Codex usage for a limited time through May 31st so that Pro $100 subscribers get up to 10x usage of ChatGPT Plus on Codex to build your most ambitious ideas.

@randombuildr @OpenAI Claude code is stingy now

@OpenAI If people are ready to spend $100, then they will choose Claude Code only.

@pmarca It's inevitable though. The demand for intelligence far outpaces the supply of hardware. That shouldn't be a surprise, that said, energy itself is the limiting factor

It’s an unfortunate time for the industry to run out of every single important component in the supply chain.