Claude: NEW AI Operating System is INSANE! artwork

Claude: NEW AI Operating System is INSANE!

AI News Today | Julian Goldie Podcast

June 2, 2026

Claude + Agent OS: The Command Center That Fixes AI Memory, Chaos, and BusyworkThe script explains why using Claude (or any single AI) feels frustrating because it forgets context when you return, forcing you to start over, and contrasts this with an “agentic operating system” that puts all your AI...
Speakers: Julian Goldie
**Julian Goldie** (0:00)
Claude Plus, Agent OS is absolutely insane. And once you see this, you can't really unsee it, because the problem nobody talks about is like, you opened Claude, you know, you ask it to do something. If we give you an example right here, it gives you a great answer. You close the tab, the next day you come back, and Claude has no idea about anything, about what you've created, what you were doing. You kind of just started from scratch again. The memory isn't very good, and you start over and over again every single day. Now picture this instead. So you open one screen, Claude is right there, but so is every other AI agent you've ever created, and everything else that you've done. That is Claude with an agent operating system, because you've got all of your agents ready to go. You got Hermes, OpenClaude, Claude, FreeClaude code, all of them on one dashboard, all sharing the same memory, all knowing what you did yesterday, all ready to work. That is an agent operating system, and it changes everything. Let me show you why this matters so much. Think about your computer for a second.
Your laptop has an operating system, right? Could be Windows, could be Mac, and without it, your laptop is just a brick. The screen, the keyboard, the apps, none of them actually talk to each other. The operating system is a thing that makes work together. It's a boss, it's mission control. Now, here's the thing that most people miss. AI agents are the same. Claude is powerful, like really powerful, insanely powerful, probably one of the most powerful tools that humankind has ever had. But, if on its own, it's just like an app with an operating, with no operating system, it can't remember, it can't talk to you over tools, it can't really track what matters to you. And the old way was simple, right? So you had one AI, maybe you were going to chat to EBT and you typed in a box, you got an answer, it was done. The new way is very different because you have a whole team of AI agents and they need a place to live, a place to work together, a place to remember. They need an operating system, they need an agent OS and that is what I've built as you can see right here. I call this a command center because that's exactly what it is. It's one screen where you sit like a captain on a ship and every other AI agent response to you. You can see what they're doing, you give them jobs, you watch them work in real time.
The cool thing about that is we can ask Hermes to do something and then we could also ask Claude to do something at the same time.
So most people are still flying blind. Ten tabs open, Claude wants some other tool or another. Copy this, paste that, lose track, start over. It's absolute chaos. The command center kills that chaos because everything is in one place. Now, let me break down what's actually inside this thing because this is where it gets really powerful. So first of all, you've got the mission control. This is the front page. And when you open it, you see all your agents line up, right? Are they online? What are they doing right now? It's like a dashboard in a car, one glance and you know everything. And then each agent gets its own room. Claude has room. You can chat with it right there. Next to the chat, there's a workspace. So when Claude builds something, that could be a page, an image, a video, it shows up right there. You click it, it plays. You don't go hunting through folders. It's all right there. If we want to check out the podcasts that we've built, we can check them out here. If we want to check out, for example, the images we've built, here we go, right? If we want to check out some videos we've created, no problem. Everything is all there, ready to go in one beautiful place, right? And the great thing about this as well, you can preview everything in real time as well.
Now, next to the chat, there's a workspace, so you've got it right here. This is where you can preview everything. And also, you've got the part that I absolutely love. This is one of my favorite parts, which is the shared memory system. Now, the way this works is we actually build something called Obsidian into the agent operating system. And this is a thing, right? The memory is the difference between a tool and a teammate. A tool forgets you the second you walk away. A teammate remembers. A teammate picks up where you left off. That's what shared memory gives you. The quick question for you, right? How many times have you explained the same thing to an AI over and over? Your business, your customers, your goals, every single time from scratch. And it's pretty exhausting, right? So with the agent operating system, you explain it once and it remembers forever. And I've actually set up a system since I day I prefer Boardroom where my agents take notes on me autonomously and the user and then documented inside my memory vault. So I have all my memory stored and then I can come back to them and we're ready to go.

8 more minutes of transcript below

Feed this to your agent

Try it now — copy, paste, done:

curl -H "x-api-key: pt_demo" \
  https://spoken.md/transcripts/1000651996090

Works with Claude, ChatGPT, Cursor, and any agent that makes HTTP calls.

From $0.10 per transcript. No subscription. Credits never expire.

Using your own key:

curl -H "x-api-key: YOUR_KEY" \
  https://spoken.md/transcripts/1000770833574