In this conversational podcast episode, Eric Siu and Neil Patel discuss the emerging problem of "AI slop grenades" – low‑quality, AI‑generated work that is handed off without review, causing extra work and subpar results. They trace the term to Tobi Lütke of Shopify and illustrate how AI can both reveal lazy employees and amplify the output of skilled workers. The hosts describe how such slop harms marketing effectiveness, leading to generic copy, banner blindness, and poor conversion rates. They reference Mira Murati (Thinking Machines) and SpaceX as examples of high‑performance cultures that reject mediocrity. To combat slop grenades, they propose a tiered accountability system (verbal warning, written warnings, paper trail, possible termination) and discuss practical enforcement in remote settings. Throughout, they use fitness and growth analogies to stress mindset, continuous learning, and personal responsibility in an AI‑augmented workplace.
Chapters
All right, Neil, let me tell you something that's been on my mind. So, have you heard of this concept known as slop grenades?
Slop grenades?
Yeah. Okay, so you know what AI slop is, right?
Yeah.
Okay, so Tobi Lütke from Shopify has been talking about this concept known as slop grenades. And so a slop grenade is, let's say, Neil, I do a presentation deck, okay? I don't review it. I send it to you. Because I didn't review it, you end up doing more work and it ends up being a piece of crap. I have then given you a grenade and it's exploded on you.
But why would you redo it? Why wouldn't you just tell the person you're fired or tell them to fix it themselves?
Well, yeah, I mean, easier said than done, right? Why don't you try doing that, right? So, so let's, let's say this, okay? So the thing is, people have been getting, you know, obviously a lot of these, these magazines like Wall Street Journal will spin it in a different way. I don't know if it's Wall Street Journal, but another magazine did this, right? They're like, you know, years ago, Toby Lutke was like, everyone's got to be using AI, and now today he's talking about slop grenades. And I'm like, dude, I commented on the Instagram post because I was so annoyed. I was like, dude, 2 thoughts can exist in the same world. You can expect people to use AI, but expect them to use it in a smart way, right? And so my challenge has been more and more people now are using like these bots, right? They're using GrokBot, and, you know, we get GrokBot and we get Claude and things like like that for people. And then what happens is when they actually— when they leave a— they send a file to me or whatever, or they want me to review something, it clearly is done by AI and they haven't reviewed it. And that just drives me absolutely insane. So one thing we should talk about, Neil, is how we can go about countering this, because I have some thoughts on how we can counter it. And I tried testing the team too. I threw it into the channel. I was like, hey guys, what are some things we can do about slop grenades, right? Yeah. And the way everyone answered was like anything but solving the problem. And so what do you think?
So let's actually go back a little bit. I hate how the media is twisting what Toby is saying because when he's talking about people need to use AI, it's great. He never once said people should use AI to do mediocre work and just push it out. That was never his intention when he was talking about, or any entrepreneur or leader who says AI is great. Second, this is a huge problem in which, and you've said this to me so many times and I really agree with this statement, AI really does show you who's lazy, like just doesn't want to do the work and they don't care to do it.
He was going to say stupid, guys. He was going to say stupid.
No, I was going to say the F word, but I was like, no, I don't want to cuss. It really does. And then at the same time, it shows you who's really amazing. Now, your amazing people, you should already know they're amazing. The lazy people, some of them hide under the radar, especially if you have a lot of people. If you have a lot of good middle managers, you can get rid of some of them, but still, it's impossible when you scale up to thousands of people to not have lazy people in your organization. I don't care if you're Facebook or Microsoft or Shopify. They all have some sort of lazy people, even if it's 1% or a fraction of a percent. Everyone does at scale. But it's really coming out and people are doing a lot of damage with AI. Like, I see it firsthand. It doesn't mean that I'm anti-AI, and I know you're not anti-AI by any means. It's just people abuse it. They're just like, oh cool. And we have a mutual friend. You know this friend, he worked for you for a little bit. Yeah. The guy is just like, oh, let me just crank out all this. Look how much I can do. Look how good I'm at marketing. And I tell him this, by the way, so I'm not saying it behind his back. But he's like, look how powerful I am because of AI. I'm like, no, you're having AI do your tasks that you don't know how to do and you're not good at. You don't really know if AI is doing a good job or not. So then when you implement some of this stuff, what's happening is, is you're not getting the results in marketing and you're blaming other things. when the reality is you don't have domain expertise within this marketing strategy, like whether it's paid ads or geo or email marketing or whatever the line item may be, and you're just producing mediocre, average, or subpar work. Doesn't matter how much you're paying for Claude and OpenAI, the results just suck. And then you're blaming the organization and other people because you never took the time to really learn That marketing skill set and get really good at it so you can dominate.
Yep. So I have some thoughts on how to counteract this, right? But before we give some thoughts on this, there's a reason why— I forgot what her name is, but she was one of the— she worked at OpenAI early, but she now has a company called Thinking Machines. And this is exactly what it is, guys.
Not just rolling— I mean, her whole— whatever her— she was ex-CTO.
Yeah, I forgot what her name— Maria or something?
No, I think she's Iranian. That doesn't sound like Maria.
Look up Thinking Machines CEO.
Mira Murati.
There we go. Mira Murati, right? Exactly. So, and thank you for that. So guys, it's not just machines, it's thinking machines. And we've said this before on this podcast, right? AI is an intelligence amplifier, and that comes from our mutual friend Syed. So if you're smart already, it is revealed. If you're not so smart, it is revealed as well, right? All will be revealed. You can see everything. And Neil and I were talking about this on the phone earlier this week about this. But what reveals it even more is, I remember I was kind of taking a look at what different people on the team were doing. Let's use sales as an example, right? So there's a period of time where someone on the sales team was just sending Claude-designed stuff, right? And it was all Claude-designed. It clearly looked like it was Claude-designed. But the worst part about AI, when you look at decks, Neil, is it's like 3 ways in. It's like the headline doesn't make any sense and it makes you feel like an idiot, right? Yeah. 3 ways in the door. The door was broken before it was even opened or something like that, right? And I'm like, am I stupid or is this thing stupid, right? It like, it starts— you start to question your own intelligence as you're reading it. Um, and so whenever you feel like you're kind of in the wilderness with this stuff, it actually means someone didn't review it, right? And by the way, if someone's using default Claude design, like remember, uh, Neil, I spoke at an event last week. I don't know if I told you this, but everyone including myself was using Claude design. You know how I could tell? There's like a little pill on this thing. There's like the little italics that it has. And then it has a stupid like, you know, it's not this, it's that thing. And there's like 3 columns. I'm like, guys, like, there's no differentiator here and you're all looking like idiots, including myself. All right. So I wanted to take a moment to tell you about my podcast co-host Neil's agency called NP Digital. And they work with a whole host of global companies or global organizations. Also, Neil has SEO tools such as Ubersuggest and AnswerThePublic. All you have to do is go to npdigital.com to learn more, and we'll see you on the other side.
Dude, totally agree. And on a quick side note, to go back a little bit, Mira is Albanian, not— she's not from Iran, so that's my bad. But dude, you bring up another good point about AI, and I always talk about this when I go to conferences. Everyone these days, when I ask who's using AI, everyone pretty much raises their hand. Not because they're doing geo or anything like that. They're using AI to actually get their work done. Small portion are doing geo and AEO type of stuff. Well, I'm telling them all because, you know, like when you're in a big room and you just ask everyone to be transparent, a lot of people genuinely are transparent. I'm like, come on, don't be shy. And few people raise their hand, then more and more. I'm like, how many of you guys use AI? Almost everyone always raises their hand. Okay, the people who aren't raising their hand want to raise their hand, they're just too lazy. And now the people who are raising their hand, I always ask them, how many times do you modify this stuff? And then I ask them, how many of you guys actually modify the work? Almost everyone raises their hand when I ask them that. Okay? And then I say, how many of you guys modify the work more than 10% of the time? How many hands do you think are usually raised?
Should be all of them still, or zero.
Less than 5%.
Yeah, they don't continue to modify, which you're reinforcing this point.
Yeah, so like they say they modify, but they modify maybe 1 out of 20 times. All right, that, that's not enough. That really isn't. So you're going to have all this me too stuff, and when I say me too, I'm not talking about the Me Too movement. I'm talking about lookalike stuff like Where's Waldo, where everyone just blends in together. And then they're like, why aren't my marketing campaigns working? Why aren't our conversions going up? Well, if everyone's using the same exact marketing copy or similar marketing copy, similar conversion tactics, similar messaging, similar flows, it doesn't convert. It's because people get numb to it. It's just like when the first banner ad came out, it had a massive click-through rate. It was amazing. People were like, wow, banners are effective. Why aren't they effective anymore? Everyone sees them so often that we all have banner blindness. That's what happens when you just use AI to just regurgitate and create the same stuff that everyone else is doing.
And guys, marketing is the act of bringing people to the point of sale, right? Which means you actually have to stand out. And so back in the day when Facebook first came out with Facebook ads, people would do like red borders around the ads and then they would get a very high click-through rate. Or people would actually have like, and this wouldn't work. I mean, sex sells at the end of the day, right? But people put like a really beautiful girl or whatever, and then the click-through rate would skyrocket. But let's come back to how to counteract this, Neil. So And then I actually have a point on talking about a company that doesn't tolerate bozos as much, right? But let's come back to how to counteract this real quick. So my thought is this. I read a book a long time ago. I had to be in my early 20s, right? And the book that stuck with me, I don't know if it was Never Eat Alone or whatever, or How to Win Friends and Influence People. I don't know which one it was, but the whole idea was this. And this story is now coming back to me. I remember there's this one story where it's like, Neil, imagine if you were my boss, right? Imagine if I turned in a piece of work before you would review it, you would say, Eric, and you would look at me intently and you'd say, is this your best piece of work? And if I said no, you would push it back to me and say, bring it back to me when it's your best piece of work. And then maybe I come back to you a week or two later and I bring you the work again and you say, Eric, is this your best piece of work? And if it's no, I have to go back and do it, right? So what I'm thinking about doing now, because again, I put this out to the team, I was hoping someone would come up with a good solution here, right? But everyone's kind of going around it. I'm like, you know what? Here's what we're going to do. If you send me a slop grenade and I have to review it, guess what? I'm not going to review your work for another 7 business days. Okay? And if you miss the deadline, it is your fault at the end of the day, right? And you will also get a warning as well, right? And if you continue to keep sending slop grenades, guess what? Those warnings are going to add up. We're going to have a paper trail for it, and it's not going to be a good thing for the long term, right? And so I'm not saying this is the solution, Neil. I'm just saying let's just brainstorm a little bit on possible solutions. No, no, no. Now the easiest solution is you're just fired, right? That would be the easiest solution.
Yeah, because the problem is, it's like, if I say, let's say you and I work together, and if I said, Eric, is this your best piece of work? For me, that's a pain in the butt. I don't want to keep asking you if it's your best piece of work. I just expect your best piece of work to be turned in. I like the idea of the second part that you said, in which, hey, if you're producing slop grenades, the first time, You'll get a verbal warning. The second time you get written up. The third time, you know, you'll get written up again. The fourth time you'll be terminated. And we'll have a discussion through this. And if you genuinely believe it wasn't a SOP grenade, you can, you know, explain why and make your case. But we don't have time for this. It's slowing down the company. I wouldn't expect you guys to do it to me, and I would never do it to you, and your colleagues should not do it to you either, because it just makes everyone inefficient. So look, don't do it. Check your work. Be thorough. We're not kids here. No one needs to be babysat. Just if you don't do it, eventually— I mean, if you don't do it, you're fine. If you keep doing it, you will be terminated. And this is just how things are going to work. Yeah. And I think that's the easiest solution. The reason I say the first time verbal— I know I'm not in HR, so I don't know how HR works. But the reason I say first time verbal is you have a good culture. The last thing you want to do is if someone, you know, doesn't check or they have some issue, you tell them they should learn their lesson and you shouldn't have to repeat yourself again. But they know it is serious at that point.
Yeah. The only thing I would modify, Neil, is because we're in such a remote environment today, the first time it happens, I'll just send an email just so there's a paper trail there. And so it's like, hey, I'm just going to let you know through email. Yeah. And so it's not an official write-up, right? But the second and third time, I agree with that. And here's the thing, like I was thinking about this the other day, right? A lot of people, like if you expect to do great things, it doesn't come from having a peaceful boss or like a peaceful organization, right? Growth naturally comes from being uncomfortable, whether it's lifting weights or whether it's, you know, having to wake up early and do things that normal people wouldn't do, right? And so when I think about an organization that maybe doesn't have as many bozos, Neil, when I've talked to a couple people that work at this company, It's SpaceX, right? Well, the thing with SpaceX is like, you get pushed, you push, you push. At a certain point, people can't take anymore and they leave, right? Which is why you see people that have worked there for like 10, 15 years, like they tried to come back and they can't anymore because they're in a different phase of their life. So I think SpaceX is actually a really good counterexample of an organization that just doesn't accept it. And again, like it comes from high standards. You're like, you're not going to actually work at a really cool company where everything is pleasant, right? Nothing great comes from just sitting and chilling.
Dude, and I love the analogy you gave about— I think you mentioned working out there and pushing hard. You know Lloyd, right? Lloyd's wife is a Stanford doctor, really smart. She posts quite a bit of Instagram content. And I love using that analogy when it comes to weightlifting and working out. A lot of people don't build muscle, as much muscle as they want. I'm not the I'm not a buffest guy by any means. I'm underweight, sadly. I have the opposite problem where even if I don't get hungry, so I tend not to— what is it called— gain weight. It was funny because when I was dating my wife, this was a long, long time ago before we were married, hence I was dating her at the time. She would be like, if I could package up the gene inside you that just makes you never hungry and sell it, we would be rich. And then years and years later, Ozempic came out.
There it is. Yeah, see, if you read— if you— it grows faster than AI.
That does grow faster than AI. And I was listening to Lloyd's wife on Instagram, and she was breaking down a study, a medical study, in which people who really build muscle push themselves so hard— they don't just do 10 reps or 12 reps, they're going until failure, or close enough to failure in which their body can't push anymore, because it It tells their body, hey, you need to build stronger muscles or however it works biologically. And it's like, just think about that. To get results, you're pushing yourself to the extreme. You and I have been entrepreneurs for a long time. We've been marketers for a long time. When we want things to grow, we push ourselves extremely hard and we go through a lot of ups and downs. You're not giving up and you're just mentally exhausted sometimes. I don't think either of us really get physically exhausted. from our work, but mentally you're getting exhausted and you're just pushing and pushing and pushing. And if people don't have that mentality of just going and going hard and at the same time learning from their failures and what's not working and adapting, then I just feel most of those people aren't meant to do well in a corporate world unless you want to be entry level.
Yeah, the saying is— go ahead, finish your thought.
And I was just saying entry level, a lot of them are going to be replaced by AI in my opinion.
Yeah. And here's what I'll say. I would say this. I mean, if you're entry level and you've got the drive and you've got the good relationship with curiosity and you love learning, you're going to be just fine, right? Because remember, I've talked about the 15-year-olds, the 17-year-olds and all that. I think again, it's a mindset thing. Like earlier this week, Neil, I talked to a 64-year-old to potentially join the team and this guy spent $40 grand. Keep in mind, he's an SEO, okay? And he makes probably $130 grand a year or something like that. So it's like a good salary, right? But here's the thing, Neil, he spent $40 grand on his own racks, right? For AI. I'm like, why did you spend $40 grand? He's like, because you got to. And I'm like, but what, who told you to buy it? He's like, I just told myself to buy it, right? So it doesn't matter if you're 64, it doesn't matter if you're 15, it doesn't matter if you're 17, it's a mindset thing. And I've mentioned this on this podcast before, you have to go on this journey on your own. Your boss cannot push you. to want to go work out, right? That motivation thing is something that you build on your own. And then, by the way, once you start to build the gene to work out, like physically work out, then you actually have the gene to build anything that you want, right? Most people just that work out a lot that are ripped all the time, they actually don't take that and then use it to build something that can be even greater than them. All right, well, anyway, guys, that's it for today and we will see you tomorrow.
More from Eric
Even Beverly Hills Can’t Escape These Grocery Prices #shorts #business #inflation
