Tim

Working in the next big thing

[Tim @TimPerelman]

Cancelling my ChatGPT Pro 100 subscription

"Your dot is on a break. Congrats you are in the very top users and have hit our abuse prevention limit. Check back in a bit!"

[Tim @TimPerelman]

@reach_vb "Your dot is on a break. Congrats you are in the very top users and have hit our abuse prevention limit. Check back in a bit!"
I didn't break my piggy for this. You announce improvements and deliver cuts.

[Tim @TimPerelman]

@thsottiaux And yet my Dot is "paused" to prevent "abuse"! This is the worst day since Dev day! You're not giving me more today, you're giving me LESS

[Tim @TimPerelman]

@melvindvivas @iamdemetrius_ How's grok bot usage? And how many bots can you spawn?

[Tim @TimPerelman]

@Jetpack1337369 @melvindvivas Why are you guys saying this?
I liked dots so far (except in the last day or two when it started feeling nerfed, and now it seems to be down)

[Tim @TimPerelman]

@thenanyu Dot is dead, did you know that? Who needs to know here on X?
It's not even showing in the chatgpt web app bar. And on desktop:

[Tim @TimPerelman]

@ah20im Dot is dead, did you know that? Who needs to know here on X?
It's not even showing in the chatgpt web app bar. And on desktop:

[Tim @TimPerelman]

@victornunez Victor, dot is dead
https://t.co/P1K61Q6VKX

[Tim @TimPerelman]

OpenAI's ChatGPT dot disappeared from the web UI and on the desktop app it is not responding (marks messages as read)

[Tim @TimPerelman]

OpenAI's ChatGPT dot disappeared from the web UI and on the desktop app it is not responding (marks messages as read)

[Tibo @thsottiaux]

All right, we’re locking in. Only things being worked on are simplifications, more efficiency for more usage, groundbreaking features or new models.

Sometimes you have to invest ahead of the curve, but feedback is clear that you all want things to get simpler. On it.

[Tim @TimPerelman]

Dots are nerfed

[Tim @TimPerelman]

During the last couple of days, my OpenAI dot has become considerably dumber, slower, less productive.

It hard for it to execute a full task without abandoning it, it doesn't report back, it can't execute its scheduled tasks. "triggered doesn't mean executed".

Disappointing.

[Tim @TimPerelman]

Something annoying about OpenAI's dot is that it keeps using a timezone that is not related to me. It also tends to forget the rules a lot. Takes some work to get it to run smoothly.

[Tim @TimPerelman]

@ah20im It is scary. I hadn't used Grok's or Muse's. Another "feel the AGI" moment.

I tried calling it and it didn't pick up the phone nor said why or call afterwards though.

I'm scared I'll eventually have to pay a lot for this.

[Tim @TimPerelman]

Dots are CRAZY. This is what I expected 'agent' and 'codex' to be way back when they were released.

[Tim @TimPerelman]

Switching from ChatGPT 5.6-Sol down to ChatGPT 5.5 to check if it is less nerfed

[Tim @TimPerelman]

Having huge Cortisol spikes from ChatGPT 5.6-Sol "Extra high" delivering counterfactual technical responses.

[Tim @TimPerelman]

"Our systems are thinking a bit more about this request before responding."

[Tibo @thsottiaux]

@ClaudeDevs Do not worry, we have compute

[Tim @TimPerelman]

@My_Ai_Bi All plans (or at least all Pro plans) had x2 credits for a while. The $200 plan the temporary relative x2 extended while it was the most expensive plan.

Now it is back to normal. New users are still getting 10x the $20 + more than 10x the frontier models' usage in Web chat.

[Tim @TimPerelman]

How can the "new AGI model" gpt-6.1-sol (high) be so stupid?

How can your products be so unpolished? smh

[Tim @TimPerelman]

@KSimilien @philipkiely Yes you can, with ocx opencodex