OpenAI Just Launched Dots. Here's The Truth About It. transcript

Open AAI just launched DOTS, an AI agent that never sleeps. [music] It works while you're offline. It texts you first when something actually matters, and it [music] keeps going with nobody at the keyboard. The second it dropped, I ran to my Chat GPT app and opened it up, and [music] it wasn't even available in my region yet. So, I got my hands on it another [music] way because I had to see this one for myself. Look, I've spent a year building on this exact kind of always on agent.

I've run Open Claw. I've run Hermes agent, Grock Balt, and every other agent under the sun. [music] So, I'm probably the right person to give you the honest truth about it. So, in this video, I'll show you what DOTS actually is. I'll set one up and hand it a real job on camera. and I'll tell you what OpenAI genuinely got right, where it falls apart, [music] and whether it's worth the money. Let's get into it. Okay, so what actually is dots?

Think of it like this. Regular Chat GPT waits for you. You open it, you type, it answers, then it goes quiet. A DOT doesn't wait. It's an always on agent that lives inside Chachi PT and keeps working when you're gone. It runs on ChachiPT 6 Astra. It gets its own computer in the cloud and it can work toward your goals 24/7. It plugs into over 4,000 apps and it learns how you like things from every correction you give it.

And you can reach it right inside chatbt in Slack and Teams or you can literally call it like a person. Open AAI is pitching it like an employee instead of a chatbot. So let me actually set one up. Setting it up is honestly simple. You open the chat GPT desktop app or you go through the announcement. Doesn't really matter which way you do it. You can just open the app and then you most likely find your dot at the top.

You give it a name and you pick what it looks like. I picked mine to look like a green frog because I think it looks the friendliest. And I also named mine Bob for similar reasons. Then it'll prompt you to connect your apps. And this part matters because the more you plug in, the more it can actually do for you. I gave it a job uh to actually run through the different features that it has and to make us a presentation about itself and how it compares to Hermes agent and Grockbot.

And it did. There you go. It went through and made an entire 10page presentation for us along with the differences between it Hermes agent and Grock ball. It also keeps doing something very interesting which I haven't really seen other agents do. It's actively reading through all of my documents that I gave it access to and it's highlighting potential tasks it can do without me ever prompting it to or asking it to do that.

This is instantly one of my favorite differentiators by far of dot with any other agent. I have to verbally tell them I want them to read or look into something. Dots just does it which can also be extremely creepy. Why is it snooping around my files? But yeah, that being said, that is or was literally the easiest way to connect an AI agent. All right, I'm going to be fair here because a few things are genuinely well done first.

And like I just said, it's proactive. It messages you first. You don't have to remember to check in. It surfaces things on its own. That's the part that feels extremely different from a normal chatbot or even from Hermes agent or Crockbot. Second, and this is another one I appreciate, is the safety model. By default, when it's working in the background, it's read only. It can look, it can draft, it can plan, but it can't send, post, or spend without your approval.

This is exactly how I tell people to run an agent. and OpenAI baked it right in. Third, OpenAI did something I hope more companies start doing more often, which is team mode. You can share access to certain artifacts or work folders that your DOT creates to other team members and they can use their own dots to work on one common project. I run an entire company and one point of friction we always have is we all use Hermes agent but it's extremely hard for us to collaborate on one project together.

Let me give you an example. If I create a folder I have to ask Hermes to compress it into a zip file which I have to send to my team member. or my team member has to send that to his Hermes agent and once they're done compress that too and send it to another team member. The bottom line is 24/7 AI employees aren't really efficient for teams but dots is and yeah you can call it it rings like a phone call and you just talk to it. >> Bob is taking his time to respond. >> Hey Charble.

Hi again. >> Hey Bob. So, what's on your mind tonight? >> Bob, I'm sorry. I I I I I thought you were a guy. >> Oh, my bad. I can be Bob with a dude vibe. No problem. What do you mean? >> This is creepy. I I thought Bob was a was a dude. I don't think I want to call Bob again. >> It's a nice touch, but it really isn't anything different from a GPT voice mode session you get on the regular GPT chat. But here's the part I have to be honest about.

And I know what you're thinking because I thought it as well. Isn't this just everything we already have? And honestly, kind of. Yeah. There's very little here that Grock bought or a Hermes agent isn't already doing. All the agents on the market are always on, plugged into your apps, working in the background, messaging you first. Granted, you set that up. We've already had this. Dot looks new, but under the hood, it isn't that new.

So, the magic here isn't that it's this new super agent with brand new superpowers. The magic is that OpenAI just put this in front of over a billion people at once. Now, after 24 hours of playing around with it, here's where I felt it mostly fell apart. First, and the most important is where you live. It launched in some markets and not others. I went to open it the second it dropped and it wasn't available in my region at all.

If you're in the EU like me, you're kind of used to this by now and I'm not a fan of this. Second, and this one's a kicker. The same day Open AAI handed you an Always On agent that eats at your usage, they quietly cut the $200 Pro plan in half. It used to give you 20x the usage. Now it's 10x the usage instead. And then they rolled out a brand new $500 a month plan that people are speculating is going to give you 25x the usage.

So they shrink your tank and hand you a car that burns more gas. Third, and connected to that, tasks cost you. Chatting with your dot is basically free, but the second you send it work, that comes straight out of your usage. And an always on agent grinding all day adds up fast. And fourth, it is day one. I mean, it's buggy. things drop and it will over message you until you rein it in. But I do think that will get better.

But that's where it currently is today. Now, dots didn't show up in a vacuum. This exact idea and always on agent that works for you is having a moment right now. So, let me put DOT up against the biggest always on agents currently on the market. First is Meta's recent Meta's Muse. This one is basically DOT's twin. It's always on, works on its own, asks before it does anything risky, and it gets its own isolated machine for you.

The difference is Muse lives where normal people already are, meaning inside Meta's apps and WhatsApp, and most of it is free. So, OpenAI's big paid launch already has a free lookalike from Meta that dropped just days earlier. Then there's Grockbot from XAI. This one leans harder into real work. It also gets its own cloud computer and instead of plugging into apps through connectors, it just uses them or can just use them.

Clicking and typing like a person would. You show it a task once and it repeats it. It is powerful, but there's no free tier for it. And it's aimed at the power user, not your everyday person. And then there are the two you actually own which are Hermes and OpenClaw. Both are open- source. Both run on your own machine. If you have a spare laptop, a Mac Mini, a cheap little server, both are always on and both plug into your messaging apps.

Both write and run their own skills and keep improving these. And here's the part that matters. The tools themselves are free forever. Nobody can nerf them, jack up the price, or lock you out because of your region. You pay for the model and maybe a few bucks for a server if you don't have a spare machine, and that's it. So, look at the pattern here. On one side, the polished, paid, lockedin agents, that's Dots, that's Croc.

And on the other side, the free agent. Muse hands it to you free inside of Meta. Hermes and Open Claw hand it to you free and you own the whole thing. And that reframes the entire question. DOTs isn't the only on agent. And right now it's the most expensive way to get one. So here's where I land and I'll split this because it really depends on who you are. If you already pay for Chat GPT Pro and you live inside of it all day, wired into your workflows or your work tools, this is a genuine no-brainer, at least to try out.

It's the most polished always on agent a normal person can get, and you already have it. If you're a builder or you just care about owning your own setup instead of renting it, then here's my honest take. dots didn't show me anything I can't already do with an agent that I own for free that nobody can nerf or price out from under me. That's still the move for me. Either way, the fact that OpenAI shipped this to everyone means always on agents just went mainstream and that is a good thing whichever side of it you're on.

So tell me in the comments, do you like OpenAI's dots? I'm curious where people all land on this one. And if you want to try this out for yourself, I put together a free resource with my exact dot setup, the rules I set, and the prompts to build your own. It's the first link in the description below. If you enjoyed this video, then leave a like. Or if you're new to my channel, then subscribe because I've got a ton more content like this coming your way.

Oh, and would you look at that? the algorithm gods have decided you're very likely to enjoy this next video as well. So, click it and I'll see you there.