Siu correctly identifies that reliability is the new frontier, effectively rebranding basic software stability as a profound strategic insight for the AI era. It’s a pragmatic reminder that a boring tool that works will always outlast a flashy one that breaks.
Inmersión profunda
Prerrequisito
- No hay datos disponibles.
Próximos pasos
- No hay datos disponibles.
Inmersión profunda
Is It Game Over For OpenClawAñadido:
So, is it over for OpenClaw? That is the real question because right now there are competitive agents coming along the way, but OpenClaw literally just issued an apology. The creator just issued an apology. And I'm going to tell you right now, I'm going to show you some examples of how [music] it is not working well for us right now. OpenClaw has become pretty critical to how we do business, how my team works, and they say they cannot do without it, right? So, having these autonomous agents is extremely powerful. I certainly think that Peter Steinberger has brought a lot. He's the creator of OpenClaw, and he's certainly given us this gift, this open-source gift. But the issue now is if it's unreliable, many people can't use it, right? So, I'm going to show you the apology letter that they issued just to show you kind of where where they stand right now. And I want to start with even just starting with kind of what what I'm calling out over here, okay? So, I literally just said that OpenClaw is compounding reliability issues in the last 1 to 2 months, you know, for us.
Our team doesn't bother talking with our OpenClaw agents anymore and has moved over to talking to the Hermes agents instead. We call ours Hermes. And I'm hoping Peter Steinberger and team resolve things soon, okay? So, literally this was posted and here this this week didn't do bad, about 18,000 views or so.
And you can see the the the the research team from Hermes or Hermes responded over here. And then actually their creator I I think he retweeted this. And so, over here you can actually see a bunch of people complaining. I'm so bearish when I try to stand up for OpenClaw. I'm going to put the effort into fixing it, but it's not working. Temporary setback.
You know, Hermes has unlocked another level of productivity, which we can talk about because that's that's one of the other agents that I'm using right now.
Everyone just saying they're seeing that the service is degrading, right? And you look at this, I'm losing patience over here. You know, same, I just moved over to Hermes. Hermes or bust. Just all this stuff over here, right? And so, people aren't happy. And so, what happened was Peter issued an apology over here. So, OpenClaw had a rough week. So, this is as of today, actually, as of this recording. So, you can see OpenClaw had a had a rough week. So, this basically this launch over here for April 29th made it obvious. So, they've been launching a lot of new updates recently.
You know, we're making the core smaller, they're moving optional stuff to ClawHub, which is a lot of skills that they have, I believe. And announcing LTS separately in May, okay? So, the trouble started around this patch over here.
It's obvious that nobody could pretend that this is just a few weird installs.
Gateway's got slower. Some installs got stuck in plug and dependency repair loops. Discord, Telegram, WhatsApp, and other channels behave worse than they should. People downgraded. People lost time. So, I'm going to show you a couple examples in a second, real-life examples of us working within Slack because we treat this as a multiplayer game and we have both Hermes and we have both Open Claws in there. And I will tell you right now, as as of this recording, Open Claws has really dropped in terms of its reliability. And he's apologizing over this. This is the creator of Open Claws literally apologizing. So, that sucks.
I'm sorry. Gateway Cold Path did too much work. We've been pushing Open Claws to become smaller, safer, and more infrastructure grade. You can see the language that he's using now, infrastructure grade. And I like to say for our our revenue agents that we that we sell that are on more enterprise grade software, you want to make sure that things are SSR, so secure, stable, and reliable. Again, that's secure, stable, reliable, SSR, all right? So, all the stuff he's mentioning over here.
I think the big thing is that he's calling this out over here. So, if Open Claws too founder driven, he's going to make a lot of changes all the time. It's going to be to just go, go, go, but stability isn't kept in mind, okay? So, too much release, review, packaging, and support works out with me. So, a lot of the work he was carrying himself, okay?
So, through the Open Claws Foundation with the help of Open IA, we are building a real team around the project.
They have to build a real team because if they don't do that, Hermes is going to eat their lunch, okay? So, the good news is that Hermes, when you have competition like this, that is actually a good thing, right? Cuz competition's going to bring out the best in everyone.
So, he's just saying, "Look, thank you for everyone who reported issues, paste the logs, all these things over here."
So, I think the most important thing over here is is is this right here, right? Realizing it's too founder driven, you have to build a team around this cuz if you want to build something great, you're going to have to you want to go fast, you go alone. You want to go far, you go with the team. It's going to get more secure. It's also going to get smaller, but it has to stay boringly reliable. It's exactly what I said earlier, right? Secure, stable, reliable, right? So, that is kind of the state of OpenClaw right now and I'm going to walk through how we're thinking about putting everything together because if we're going to be using, you know, Hermes, OpenClaw, Nemo Claw for our clients, it's really important like the the way we think about building things is within this this single brain, right? Having the single brain over here where you have a bunch of tools that connect to this this single brain and they're able to talk to each other, but you need reliability ultimately cuz it is if if you have this agent fleet over here where you have, you know, OpenClaw, you have Hermes and then you have your other sub agents, they need to be reliable cuz if if Alfred is working with Hermes over here and they're kind of the brain talking to each other and you can see the single brain over here, if one's not working, everything else in the system breaks, okay? So, this is how we're thinking about it overall. It's it's you know, every person should be able to get their own agent fleet that's customized and so, I'm going to show you what's happening real time right now inside of our Slack where I'm going to show you the real issues with OpenClaw.
I'm going to give you my final thoughts on where I think like is it actually over for OpenClaw or not and, you know, what what what I think the alternatives are as well. So, let's take a look at this. So, we're inside of Slack right now and this is one of our channels that we have. Okay, so Alfredo here is asking Alfred, which is our OpenClaw bot, if he's here. So, OpenClaw will respond with a emoji just to say, "Hey, like I I I saw your message, okay?" Here's the issue. You see this? 12:56 p.m., no response. 12:57 p.m., no response. 2:05 p.m., Alfredo trying to get it to respond, okay? And then 2:45, "Please, I need this information." Fail, fail, fail, fail.
I actually it's five times. Five fails over here, okay? It's just not responding, all right? And then literally with for for Vic over here, Alfred just doesn't respond. So, you see six failures over here, okay? That is just unacceptable. So, this is Slack right now. My team will try to call on these bots sometimes inside of Slack.
That is one area to look at. Now, let's go ahead and take a look at my Telegram.
All right, so now we're in my Telegram, okay? So, literally OpenClaw is building me a new page that we have for our single brain right now, okay? So, and and give it feedback, okay? So, I give the feedback. Look at this. Literally, sometimes it won't respond to me at all, all right? And then also, no response generated, please try again. So, I literally just showed you seven failures, okay? So, imagine if you're using this for yourself, for your business, you can't you can't use it right now. It's unusable, okay? Because of the reliability issues. And um if you try using it for let's say you're trying to do it for client work or you want your team to use it, you can't do that either. It's just not reliable, right?
So, where do we go from here? Like what are the alternatives? So, let me give you those alternatives. What I'm using over here, what I just talked about is Hermes. Hermes is basically they are the competitor. I think they have more stars on GitHub than Cloud Code, I believe now. So, Hermes is is extremely reliable, at least right now, okay? And we've we I've continued to add to it.
And what's cool about Hermes is that you can you can see that you can see it work, which is cool. You can also like when you do work, it creates skills for itself and remembers over time. It gets better with you. It compounds with you over time. And that's very helpful because you want something that's going to grow with you. The other piece is when I just whenever I just ask it to do something, it's going to respond very quickly. I'm just over here I'm just working on a strategy and and and just going back and forth with it. So, whenever I ask it to do something, so far it hasn't failed me once, okay? And when I think about today, when I think about the number of times in the last couple of weeks or so where where it's maybe gone down, that has not happened, right? And literally, I'm just working with it on strategy over here and that's that. And so, the way I see this is the way you want to operate in the future is you would you definitely want to have Hermes. And because Hermes is more reliable right now, you want it to be the brain, okay? Now, Hermes is the brain and then Alfred you want Alfred to be the execution arms, right? And then Alfred can be that that uh the execution leader, the fleet leader. And then you might have agents like the agent fleet that falls under Alfred and Alfred is just making sure that they're doing things, okay?
Now, ultimately, what I think is going to happen is that Open Claw is going to become a lot more reliable. I mean, this was a wake-up call for them. So, I'm not saying throw away Open Claw. And so, I mean, I think right now Open Claw is going through it a little bit. I think we all there's there's ups and downs with with having any company at all, especially when it's open company. But, now that Open AI is going to put resources behind it, and I think Peter Cyberg is a very smart person. He brought us to this this to us in the first place. I think Hermes and Open Claw will push each other. I think you should definitely be using both. And I think you should be If you're starting out right now, maybe just go with Hermes. But, eventually, I think you're going to need you're going to need both because the way I see this is that Open Claw has a lot more support, a lot more community behind it. And you There's a lot more bells and whistles you can put behind it because it has that support.
For example, I can send you a Google Meet meetings. There's a lot more integrations with it. And so, I like Open Claw for the bells and whistles, and I like Hermes for the stability because Hermes can't just do everything under the sun right now, right? And so, that being said, they're they're pretty damn close. But, if you had to start with one right now, I'd just go with Hermes. From there, like I don't want to take this away from Open Claw. I'm going to show you something real quick because literally, I deployed this using Open Claw today, okay? It's not like it's a complete piece of crap, okay? It's just gotten less stable over time. They they finally realized that. If you want an AI-native marketing firm, look no further than Single Grain. Most people right now work in the Stone Age. Not us.
When we think about using agents to supercharge our work in terms of pumping out higher-quality content creative at scale, whether it's on the ad creative side or on the AEO side, that is exactly what we do. When it comes to being proactive, thinking ahead, using the latest products to help us be more efficient so we can work on human things and we can offload the robot things to robots, that's exactly what we do. Go to singlegrain.com to learn more, and we'll see you on the other side. This is the Single Grain website, okay? Doesn't look bad. You look at this over here. One trusted brain per client. You look at the design of this, the marketing partner, a company cannot outgrow a simple chain of command. This is how our our our fleet of agents works. Vertical specialists combined into one client brain. This is This is how we're playing the game with with Single Grain, right?
You know, literally, by the way, did I need to get someone to help me design this? Nope. Did I need to get someone to help me deploy this? Nope. They didn't need to do that, right? Were there some issues with with this initially deploying it? No. But, the the form fields, all these things, passing it over to Slack, literally deploying a website, getting all these things like Look, there's actually even a logo up here, too. This was done using Alfred.
You could do it, and Alfred is our open claw chief of staff, but you can also do this using a Hermes for example, right?
So I'm saying like you absolutely need to be playing with these autonomous agents. They are an absolute Godsend. Is open claw dead? No. Is open claw struggling right now? Absolutely. Will open claw figure it out? I trust that they will. And that being said, by the way, I'm showing you the single brain site right now. If you want help building these revenue agents, running these revenue agents with security, stability, and reliability, not on open claw right now, on the most stable tech that you can have to help grow revenue faster, grow your marketing faster, then just go check out singlebrain.com and we'll see you on the other side.
Videos Relacionados
OpenHuman VS Hermes AI: Who Wins?
JulianGoldieSEO
285 views•2026-05-29
Long-Running Agents — Build an Agent That Never Forgets with Google ADK
suryakunju
142 views•2026-05-30
This computer is made from real human brain cells. And you can buy it.
Talktmsmedia
3K views•2026-05-28
BREAKING: Microsoft’s New Image Generating Model Beat Out GPT 1.5 and Nano Banana 2
aimmediahouse
122 views•2026-06-03
I Made the Same Anime Fight Scene in Every AI Video Generator
NobleGooseAnime
295 views•2026-05-30
Nvidia Bets Big On AI PCs | New Chip To Power Windows Laptops | Technology | AI Updates | N18S
cnnnews18
3K views•2026-06-01
I Tested NEW Opus 4.8 on Four Projects (Updated LLM Leaderboard)
AICodingDaily
298 views•2026-05-29
3D Platformer Update - NO CAPES
SolarLune
294 views•2026-05-30











