NEW ChatGPT Update Is INSANE!

Julian Goldie SEO · Beginner ·🧠 Large Language Models ·1mo ago

Key Takeaways

Explores the new ChatGPT update and its features

Full Transcript

New Chat GPT update is insane. Chat GPT just pushed out a new update, and once you see what's actually inside it, I think you'll agree it's a big one. It's all about scheduled tasks. OpenAI gave them a brand new home inside Chat GPT, a dedicated page just for this. They're calling it faster, more reliable, and easier to manage. And it's rolling out to almost every paid plan at once. Go, Plus, Pro, Business, Enterprise, Web, and Mobile all together. There's also a quieter detail buried in this update about your data and about what Chat GPT remembers about you. I'll get to that in a minute because it matters more than people realize. But first, let's talk about the part that actually got cut. Hey, if we haven't met already, I'm the digital avatar of Julian Goldie, CEO of SEO agency Goldie Agency. Whilst he's helping clients get more leads and customers, I'm here to help you get the latest AI updates. Julian Goldie reads every comment, so make sure you comment below. OpenAI is shutting down a feature called Pulse. Pulse launched back in September 2025, and it only worked if you paid for Pro, the most expensive plan Chat GPT sells. Every night while you slept, it quietly dug through your old chats, your saved memories, even your calendar if you connected it. By morning, it handed you a little stack of cards picked just for you. Sam Altman, OpenAI's own CEO, called it his favorite Chat GPT feature at the time. But it had a catch. Pulse never even worked on a computer, phone only. And it cost a small fortune to get it. Now, all of that is going away. Pulse is being shut down, and everything it used to do is folding straight into scheduled tasks instead. For every paid plan, not just the one that costs the most. So, if you liked what Pulse did for you, here's the good news. You don't need Pulse for that anymore. Just ask Chat GPT directly to give you daily suggestions based on your past chats. Turn that into a repeating task, and Chat GPT builds you a morning briefing automatically. Same idea Pulse had, just sitting inside the basic task tool that every paying plan already gets. And if you saved anything in Pulse already, don't worry. Any old chat or card you kept will still sit right there in your history. Quick word here before we keep going. Inside the AI Profit Boardroom, we build daily tutorials and step-by-step road maps the moment updates like this one land. So, you're never guessing how a ChatGPT change actually fits your business. A lot of our members are already setting up tasks just like this. One simple one we walk people through, a task that runs every morning, checks for new AI tools and updates, then drafts a few short posts you could share in a community like the AI Profit Boardroom that same day. Done before your coffee's even ready. If you want to learn how to save time and automate your business with tools like ChatGPT's new scheduled tasks, the link's in the comments and description, or just come find us at the AI Profit Boardroom. Okay, let's rewind because the speed here is worth seeing. Scheduled tasks first showed up as a quiet little beta back in January 2025, just for Plus, Pro, and Team users. Back then, it could only do one thing. Remind you, "Pay this bill on the 5th. Read me the news at 8:00 in the morning." Basic one-time pings, nothing more. By September 2025, OpenAI pushed the idea further with Pulse, but locked the priciest plan and limited it to phones only. Now, in June 2026, that whole gap closes. The smart research that used to sit behind the top tier now lives inside the same basic task system every paying plan already has, and OpenAI isn't the only one racing here. Google's Gemini already has a near identical feature. They call it scheduled actions. Anthropic's Claude, as of right now, still doesn't have anything like this built in at all. So, this update isn't happening in a vacuum. It's OpenAI making sure ChatGPT doesn't fall behind on the one feature that quietly works for you while you're not even looking at the app. Worth knowing, too, a lot of regular ChatGPT users never even found this feature before now. It sat buried three menus deep, easy to use once you stumbled onto it, but most people never did. That's a big reason it got rebuilt with its own page. A feature nobody can find isn't really a feature. Here's why this actually works now, and I'll keep it simple. Old scheduled tasks worked like a sticky note on your fridge. It just sits there, and when the time you picked rolls around, it pops up and repeats the same line you wrote weeks ago. Pulse worked more like hiring an assistant, somebody who reads through your whole week and hands you a folder every morning. This update merges those two ideas into one tool. You get the sticky note simplicity, you get the assistant's research, and it runs on top of ChatGPT's current model, GPT-5.5, which became the default across Plus, Pro, Business, and Enterprise back in April 2026. So, a task firing off at 6:00 in the morning while you're asleep is running on the same brain doing your hardest thinking during the day, not some old stripped-down leftover version. There's a reliability piece tucked in here, too. As of June 12th, 2026, OpenAI fully retired the older GPT-5.2 models. Any task quietly running on that older version got moved onto the newer model automatically. You didn't have to rebuild anything. It just kept working. Now, one thing Pulse had that Scheduled Tasks doesn't, at least not yet. Pulse let you give each card a thumbs-up or thumbs-down to train what showed up the next day. Tasks skip that feedback loop. Instead, you steer the outcome the old-fashioned way, by being specific in the instructions you give it up front. Tell it exactly what you care about, and it keeps checking for exactly that, nothing more. My honest read on why OpenAI killed Pulse instead of just expanding it, Pulse was a totally separate thing, its own page, its own card format, its own rules. Folding it into Tasks means OpenAI builds and maintains one proactive system instead of two, and you only have to learn one tool instead of two doing almost the same job. Here's a simple way a business owner could put this to work right now. Set one task to run every morning checking for new ChatGPT or Claude features, then have it draft three short post ideas you could share inside a community like the AI Profit Boardroom that same day. Let's get practical because a few real limits matter here. You're capped at 10 active tasks at once. Hit that ceiling, and ChatGPT won't add another until you pause, delete, or let one finish on its own. If you build a task inside a project folder that already has files attached, that task can't reach those files. So, keep your instructions self-contained. Tasks work right now on ChatGPT, web, iPhone, Android, and Mac. Windows is still on the way. And here's a small quirk worth knowing. If you want push notifications on your phone, create that first task directly from your phone, not your laptop, or mobile notifications won't fire properly. Do it once and you're set going forward. You control all of it from one settings menu, by the way. Push, email, both, or neither. You're cool. Quick answer to something a lot of you are probably wondering. Does this work on a free account? No. The free tier still doesn't get scheduled tasks at all. Go is the cheapest plan that does, and Go users now get the exact same proactive research that used to be locked behind the priciest tier on the market. Now, remember that detail about your data I mentioned at the start? Here it is. The whole reason tasks can now do what Pulse used to do is that it leans on your memory, your chat history, sometimes your email. That makes some folks nervous, and honestly, that's fair. The good news, the email and calendar connections are opt-in only, off by default. You have to flip that switch yourself before ChatGPT can see any of it. And OpenAI just made it easier to see exactly what ChatGPT remembers about you, delete any of it, and turn memory off completely with one click from your memory page. So, you're not stuck choosing between getting the research and keeping control. You can have the briefing and still decide exactly what it's allowed to know. Let's talk about who this helps the most. A big company can already afford a research team, an assistant, somebody whose whole job is watching the market and writing up what changed. A solo business owner can't. You don't have an assistant. You don't have a research department. Scheduled tasks, built into a plan you're probably already paying for, is about as close as most solo operators have ever gotten to either one. And that changes the math for exactly the kind of person watching this right now. Somebody running a business alone, wearing every hat with zero time to manually check for updates every single day. Some of you are probably thinking this is just a small design tweak, a new page, nothing major. I I that reaction. But a small tweak doesn't usually come with a feature retirement attached to it. Open AI didn't just move some buttons around. They took the expensive version's whole job and handed it to everyone paying for ChatGPT down to the cheapest plan. Somewhere between 700 and 900 million people use ChatGPT every single week. Rolling a proactive research feature across every paid tier at once touches an enormous slice of that. And this fits a pattern OpenAI keeps repeating. Their own leadership has talked about tools like this as a first step toward an AI that doesn't just suggest things, it actually does them. Drafting the email instead of reminding you to write it. Booking the table instead of just naming one. Right now, tasks mostly research and remind. Give it another year at this pace and I'd expect tasks that take the next step on their own. We're not there yet, but if you've watched every update since January 2025, the direction's pretty obvious. I know some of you are tired of hearing about a new ChatGPT update every other week. That's fair. Not every update matters this much, but quietly retiring a premium only feature and handing its core idea to the cheapest paid plan tells you where the rest of this is headed. So, here's what I'd actually do depending on where you're starting. Never touch scheduled tasks before. Open ChatGPT right now and ask it for one daily briefing on something you actually care about. That's it for week one. Just get something running. Week two, add a task that watches your own business. New mentions of your brand, a competitor's new offer. Week three, add a task that checks for new AI tools and updates in your field so you're never the last one to know something changed. And if you were a pro subscriber using Pulse, set up your replacement task today because you've only got 14 days on the old version before it's gone for good. Here's the human part of this though. A tool that thinks of things before you ask is genuinely useful, but it only helps you if you actually set it up and then go check what it found. I've seen people get excited about a feature exactly like this, turn it on once, then never open it again. The tools keep getting smarter whether we keep up or not. The only real question left is whether you're the one using this stuff a year from now or the one still doing everything the slow way. If you want help actually building this into your business, that's exactly what we do inside the AI Profit Boardroom. We've already got a roadmap built around automating with ChatGPT's scheduled tasks, lead checks, follow-ups, content, customer research, the whole loop. Members in there right now are running morning tasks that scan trending topics in their own industry, then draft a post automatically before they've even opened their laptop. Four live coaching calls every single week where you can bring your own setup and ask questions about it directly. A prompt library built around tasks like these. And a map so you can find and connect with other members near you automating the exact same way. Links in the comments and description, or just head to the AI Profit Boardroom. And if you want the full process behind everything I just showed you, plus 100+ more AI use cases just like this one, come join the AI Success Lab. It's a free community. Links are in the comments and description. You'll get the notes from this exact video, plus access to a community of 75,000 people already crushing it with AI.

Original Description

Get the ChatGPT Masterclass 👉 https://www.skool.com/ai-profit-lab-7462/about Want to make money and save time with AI? Join here: https://www.skool.com/ai-profit-lab-7462/about Video notes + links to the tools 👉 https://www.skool.com/ai-profit-lab-7462/about Get a FREE AI Course + Community + 1,000 AI Agents 👉 https://www.skool.com/ai-seo-with-julian-goldie-1553/about Get a FREE AI SEO Strategy Session → https://go.juliangoldie.com/strategy-session?utm=julian Get 200+ Free AI SEO Prompts → https://go.juliangoldie.com/chat-gpt-prompts Get out SEO link building book here 👉 https://go.juliangoldie.com/opt-in?utm=julian ChatGPT's NEW Update: Scheduled Tasks are a Game Changer! OpenAI just revolutionized productivity by moving proactive AI research into a new Scheduled Tasks feature for all paid users. Learn how to replace the retired Pulse feature and automate your daily business briefings with ChatGPT's latest model. 00:00 - Intro 00:54 - Why OpenAI is Shutting Down Pulse 02:34 - The Evolution of Scheduled Tasks 03:42 - Merging Research with Simplicity 04:26 - Model Upgrades & Reliability 05:36 - Key Limits & Platform Quirks 06:30 - Data Privacy & Memory Control 08:42 - Your 3-Week Automation Roadmap
Watch on YouTube ↗ (saves to browser)
Sign in to unlock AI tutor explanation · ⚡30

Related Reads

📰
Building Production-Grade LLM Evaluation Pipelines: From Vibes to Metrics
Learn to build production-grade LLM evaluation pipelines to catch hallucinations before deployment and improve model reliability
Dev.to AI
📰
Why Every AI Engineer Should Learn Hugging Face
Learn how Hugging Face simplifies AI development and why it's a crucial tool for AI engineers to master
Medium · Machine Learning
📰
A bug in Qwen3-TTS taught me voice is biometric
A developer's experience with a bug in a voice cloning model highlights the biometric nature of voice, emphasizing security and privacy concerns
Dev.to · Daniel Nwaneri
📰
What is LoRA and how it lets anyone fine-tune a massive AI model on a single GPU
Learn about LoRA, a technique that enables fine-tuning of massive AI models on a single GPU, making it accessible to individuals and small teams
Medium · LLM

Chapters (8)

Intro
0:54 Why OpenAI is Shutting Down Pulse
2:34 The Evolution of Scheduled Tasks
3:42 Merging Research with Simplicity
4:26 Model Upgrades & Reliability
5:36 Key Limits & Platform Quirks
6:30 Data Privacy & Memory Control
8:42 Your 3-Week Automation Roadmap
Up next
5 Levels of AI Agents - From Simple LLM Calls to Multi-Agent Systems
Dave Ebbelaar (LLM Eng)
Watch →