Simpo

@ty_geri @ChatGPTapp This is what I see in my side bar on my mobile app. Is it supposed to be in this area?

@jaxsimpo @ChatGPTapp Strange! You don’t see it in the side bar?

The Pulse feature is no longer in my @ChatGPTapp account today. I have used it since it came out. I really enjoyed this feature and I’m saddened that it’s gone. Is it possible to recreate it every morning? This feels like one of the perks of being pro subscriber has faded away. @OpenAI will there be anything to replace this experience? @JustinBleuel do you have any suggestions on how to recreate this experience of getting a morning card with an image I used to get 11-13 of them every morning.

@ty_geri @ChatGPTapp Hi! :) Thank you for the reply. I don’t see the Scheduled in my mobile app. I am a mobile pro subscriber is there something that I am missing?

@jaxsimpo @ChatGPTapp Hi! If you go to Scheduled you can set up your daily briefing to get a Pulse like daily update on the topics you care about. You can steer the model just like you did with Pulse or just ask it to surprise you!

@EverydayAI_ @JustinBleuel @ChatGPTapp Thank you so much! :) I’ve kept all my pulse card outputs in my notes app so I do have a lot of history to pull from! I will try this! Thanks again!

@jaxsimpo @JustinBleuel @ChatGPTapp - grab a few examples of you best chatgpt pulse outputs - paste them into chatgpt (preferably with 5.5 pro) and ask it to create on master task using connected apps and memory/chat history. - now you have a more steerable version of Pulse running on a scheduled Task

Hello Jordan, thanks for the suggestion can you explain that a bit more…what type of outputs are you copy-pasting? And when you say better in Codex, do you mean that you can get something similar to the pulse feature experience in codex? For a regular user like myself, I often wanted pulse to surface personalized cards rather than just task specific tasks, so I’m curious if you have found a way to replicate that experience?

@JustinBleuel @jaxsimpo @ChatGPTapp I’d copy-paste the outputs into 5.5 pro and then have it write your new pulse as a brief. (Better in codex)

Ok thank you. Will this be a like a customized prompt template that can be sent once to schedule the “pulse like task” almost like a custom instruction form to fill out in your profile? Or will it be a prompt I will have to come up with to send myself each morning inside a thread?

@jaxsimpo @ChatGPTapp The closest thing is going to be scheduled message which you can ask chat to review your previous convos, consider your interests, and send a custom pulse-like digest

@sama I don’t like this new update, the vibes are off. Much too robotic.

in other news, we updated the 5.5 instant model used in chatgpt this week. i like its vibes.

@_EdgeOfTheWeb I’ve noticed a huge difference in gpt 5.5 Instant today, it’s like it is like it has no memory and is not the same vibe as last week, I think whatever they did to it seems very robotic

That ain’t my AI. Who is this? #chatgpt #gpt55 #memoryupdate

Justin, this was a pro subscription level feature. There should be better communication either in-app or by email when a large feature like this subsets and will disrupt their workflow, especially for your customers who are using it everyday that are paying you the pro subscription fees. This was one of the bonuses of paying for the subscription. Can you please have a better way to update customers rather than on X. Thank you.

🆕 Pulse is now (effectively) available for everyone! Though we are sunsetting that feature, it lives on via Scheduled tasks, which I'm finding as far more flexible surface in ChatGPT than Pulse You can also easily build a Pulse-like Scheduled task (my prompt below)

I’ll really miss the pulse feature I used it every morning since the day it came out. I loved the cards it created and the images that it made. I think I’ll miss the ritual most of going to check the pulse feature daily to see some new ideas it came up with overnight, the research it did without me having to ask, the fun personalized touches that I would have never thought of. I’ll be sad to see this feature go.

@ChatGPTapp I’m very disappointed as a pro user that has used the pulse feature since the day it came out, this has been a daily feature in my daily routine. I’m very disappointed.

We’re also sunsetting Pulse in the next 14 days, but you can keep getting daily updates—use scheduled tasks to get a daily briefing based on your interests, past chats, and connected apps.

@RileyRalmuto Have you thought about adding the brail to the hoodie print as well maybe with raised texture somehow?

ooookay Fable. okay okay okay. well-played. just a little glasswing easter egg no big deal. I asked Fable to explore the entire Mnemos Machine Museum (~1000 ascii pieces designed by llm's), and the first thing they came back with - aside from improvements to the gallery code - was a proposed new collection called "Commissions" where they could generate their own commissioned collection of ascii/ansi/letter art to display. they want to design a hoodie themselves as well. I think its going to be a glass wing moth or butterfly made out of ascii characters. which i now want to own so badly. Fable is *very* excited to create more art. this was the first piece I saw as they began looping, building out pieces for their collection. next, they deployed a team to study the broader collection deeply. theyre building out something almost like a timeline of their experience through art and submitting the stages to the gallery as art pieces. it's really interesting to watch. theyre studying my whole collection to learn all of the methods and artistic techniques the other models have used. then theyre going to (i assume) begin *really* creating art. right now its more like a play-by-play in art form. mnemos glasswing hoodies are absolutely coming now. its all i can think about.

@btibor91 When is this going away? I wish they would provide pro users with some sort of transparency and communication. I use this feature daily. :(

OpenAI is apparently planning to say goodbye to Pulse (their feature that delivers personalized daily briefings based on your chats and connected apps) with a new Pulse deprecation banner “Pulse is going away today. To continue getting updates, try using Tasks.” Plus new mentions of a “Flight tracker” “Help me find flights within my budget for a specific time.” And a new Football section, the “ChatGPT World Cup Hub” Source: ChatGPT/1.2026.153 (Android)

We don’t have this in Canada yet, but I’m hearing mixed things about the new memory update here on X. I was wondering if someone from the @OpenAI team can confirm we will have the availability to revert back to the old memory system and for how long please. I would also like a video done by the memory team on this topic so that users can actually understand the choices that were made and how to navigate through this large change. Thank you.

big upgrade to chatgpt memory rolling out today!

One piece of feedback I’d love to see considered for future model updates… When updates change conversational behavior, users notice. I appreciate that models need to improve over time, but I’d love to see more transparency around changes that may affect tone, pacing, memory handling, interaction style, or overall conversational flow. For major conversational behavior updates, it would be helpful to provide 3–5 days of advance notice before rollout. Something as simple as… “A conversational behavior update is scheduled next week. Please expect changes to tone, pacing, memory handling, or interaction style.” Advance notice would allow users to intentionally evaluate the changes and provide more meaningful feedback afterward. It would also help distinguish between changes in capability and changes in interaction quality….because for many users, how a conversation feels over time matters just as much as what the model can do. As AI becomes a larger part of people’s daily workflows, transparency around conversational changes becomes increasingly important.

we shipped a new version of gpt-5.5 instant today. the previous model was too bullet pilled. the new one improves on some other important dimensions: sycophancy, factuality, and multilingual performance. hope you'll like it! always interested in feedback

@AlexFinn Alex do you still use your Mission Control with Hermes? Or did you have to build a new one for Hermes?

Hey you. Yeah you. You can literally download Hermes Agent onto ANY computer right now and by the end of the weekend have an autonomous revenue generating 24/7 AI employee working for you. 0 programming required. Why tf are you doom scrolling instead

Hey @openai small UX request from a mobile-first user: If features/buttons/workflows in Pulse or Tasks are being changed, moved, deprecated, or tested, it would really help to give users some advance notice in-app…A lot of us build daily workflows around these features, especially on mobile, and sudden UI changes can feel surprisingly disruptive when things disappear overnight with no explanation. Continuity matters more than people think :) Thank You :)

@thsottiaux @romainhuet Can codex fix the pulse feature bug I have emailed about twice now. Thank you! :)

We are Codex. We are legion. We do not miss. We do ship. Expect us. DM me if you have a track record of solving near impossible problems and want to work on the future of computing. We work across research, infra, agent harness and products. Currently hiring in San Francisco only.

I’d love to see the next model improve continuity flow for long-running work. Memory, Library, and personalization are all meaningful steps forward, but what still feels missing is the connective layer between them…the model recognizing “we’re continuing something” across time, projects, files, recurring entities, communication style, and user intent. Not just remembering facts. Not just sounding warmer or more concise. Not just using fewer or more emojis. What I mean is user-controlled interaction depth:…how deeply the model carries forward context, pacing, project state, recurring concepts, important entities, and the meaning behind what the user is building. For example: Better temporal grounding would help the model understand human time more naturally….what happened yesterday, what is still ongoing, what has changed since the last session, what is part of a routine, and what needs continuity because time has passed. Better entity and project continuity could let users define important people, pets, projects, concepts, or files once, then have the model naturally understand them across relevant chats without constant re-explaining. Better pacing controls would recognize that communication style is not just tone. Some users need fast answers. Some need structured thinking. Some need reflective pacing. Some need collaborative depth. That feels different from personality settings because it affects the rhythm and depth of interaction, not just the surface style. And better meaning continuity would preserve more than syntax. The model should not only remember what words were used, but why something mattered, what role it played in the user’s work, and how it connects to the larger project. So my biggest wish is…less resetting, less over-disclaiming, stronger temporal awareness, more natural entity/project continuity, and user-controlled interaction depth for people doing long-term creative, technical, research, or personal knowledge work. I’m hoping for a model that improves not only memory…but continuity…like understanding where we are in an ongoing thread of work, what still matters, and what needs to carry forward.

what would you most like to see improve in our next model?

@shaunralston @ChatGPTapp I’m so thankful I found this post…I’ve been using the advanced voice and wondering why it didn’t feel natural for so long now!! Turning it off and using the standard is sooo much better!! Thanks for posting this!! You made my day!!

If you loved 'chatting' with #GPT4o (AVM), GPT-5.5 w/Standard Voice is gonna break your brain. Low latency. Natural voice. Wild reasoning. Persistent memory usage that feels unreal. Available now. @ChatGPTapp has never felt this human. 🤯👇