I'm going to jump in, Phil Bell. Okay. Hey, Shep. Hey, thanks for coming in today. You're welcome. I appreciate it. Phil, you are the president and CEO of SMA, right? Yep. And so let's just start at the most basic level, which is tell me what SMA stands for. Sure. It stands for Steel Manufacturers Association, Shep. Okay. All right. And tell me what SMA does. Well, our mission is pretty simple. We represent domestic steel producers, and what we're trying to do is to help them safely and sustainably build their businesses, help their employees build their lives, and help the communities where they operate build their future. We represent them in Washington, D.C., and we are advocating on their behalf because a strong country needs a strong steel industry, and so that's what the SMA is trying to do. That's a lot of ground. So let's wind that back a little bit, which is you gave me a couple bullet points on what it is you guys do. So you represent producers. Yes. So who are producers that you guys, you know, top three, top five people? Yeah, we represent domestic steel producers. We're the largest steel industry trade association in the U.S. So we represent the new cores of the world. the steel dynamics, commercial metals, iconic American companies like that. Also, we represent foreign-owned steel companies that have U.S. operations, so like Gerdau, SSAB, Valorac, and many, many others. And in terms of the amount of steel-making capacity, the number of employees, the amount of capex that's being invested, by any relevant metric, we're the largest steel industry association in the U.S. Fair to say for anyone who's not familiar with these companies, you're representing like Ford, Chevy, the biggest that are out there and that are really bedrock companies, meaning that these companies, any room that somebody's sitting in right now, one of those companies. has a product that was made by them likely in that room. You're absolutely right, Shep. You know, steel is everywhere. It's ubiquitous. You know, you are using steel when you wake up in the morning to make breakfast, when you get in your car to drive to work, when you're going across that bridge to your office building. There's steel everywhere, and those companies make it for all of those various applications. And so when... You say you represent and work with these companies. Maybe give me an example of how you guys may be working together to get done what you guys are trying to get done. Okay, that's a good question. At the onset, I said a strong country needs a strong steel industry. So one of the things that we're advocating for our members are... Free and fair trade policies. And as part of that, the Section 232 tariffs are a big piece. The Section 232 tariffs are working, and they've been enormously successful. Unfairly traded imports are down. Steel production is up. A lot of people don't know this, Shep. For the first time in almost 30 years, the United States has surpassed Japan to become the third largest steel producer in the world. That's an amazing accomplishment because when I walk around the country, I still hear people say, oh, I didn't even know we had a steel industry in the U.S. Wow, okay. And if you want to, you know, really get me riled up, that's all you need to say to me. Well, I understand that. And maybe thinking about it from... You know, I'm in the steel industry, so I get it. But I think that there is perhaps a feeling out there that our production facilities are, you know, in such dire need of capex and that they're not as productive as they once were. But, you know, maybe there is a cohort of our facilities that maybe that is true. But if we're talking about being the third largest producer, You know, we clearly have a cohort for which that is totally untrue. So what is our manufacturing distribution look like as far as there? I know that we have some sort of old stodgier places that could use some help. What does sort of our landscape look like, like our physical infrastructure for making steel? Yeah, it's actually pretty good in the United States. Our steel industry is the cleanest, most efficient, and most productive in the world. Just to give you an example, 50 years ago, Shep, it took almost 10 man hours to make a ton of steel. You fast forward to today, and even if you average in those companies that do need a little bit of help, It's about one man hour per ton. And then for benchmark producers like the Nucors, the Big Rivers, SDI, CMCs of the world, they can make a ton of steel for about 0.5 man hours a ton. That's efficiency. And quite frankly, the rest of the world is working very hard to make their steel industry look like ours. And it's due primarily to the fact that we have about 70% to 75% EAF production. It's due to the fact that we recycle ferrous scrap. So we are basically taking something that was steel and making it into a new steel product. So it's a really dynamic landscape that we're in, and we're very glad to be part of it. this time so that's remarkable so let me jump back into something you talked about earlier which is i think something that is perhaps more misunderstood than it is understood tariffs so i i know that you know when we're in the political realm that means one thing And when we're in sort of like this kind of more practitioner's realm, it means something else. Sure. So, you know, the political side is not particularly interesting to me. What's interesting to me is like how what are the real effects that you're seeing in like, you know, when the rubber meets the road? Is this something that is you believe is helping our manufacturers and is putting us on the. on the path that we hope, which is free and fair trade? That's a great question. So you're going to need to indulge me a little bit, Shep, because it's a pretty long answer. I think when you look at the current tariff regime we're in, you've got to understand several things. The 232 tariffs were put in place under the first Trump administration. It was an executive order. And ironically, they have bipartisan support. President Biden could have done away with them with the stroke of a pen, and he chose not to. He kept them in place. So we've been in some type of tariff environment, the steel supply chain, for about eight or nine years, okay? And there was not a rash of bankruptcies. There was not, you know, all of these shortages of material. So that tells you something there. And the reason that we have the tariffs was to stop unfair trade. I think people forget that the United States has one of the freest and most open markets in the world. And so if you're a foreign country, you don't have a right to sell your steel here. It's a privilege to sell your steel in the United States. And if you're going to do it, you need to do it fairly. You need to follow the law. You need to do it in an environmentally and sustainable way. And that's what the tariffs have done. Also, it's caused my members. to invest over $30 billion to build new facilities, offer new product categories, to help high-end manufacturers and to make sure that they don't have to look elsewhere to get that steel. Also, I don't look at it as the price is going up. I see them going back to what they would have been if market principles were followed by all of the steel traders. So I think that's really important. At the end of 2024, Shep, only about 18% of all steel imports were even subject to the 232 tariffs. A manufacturer, an end user, could get steel domestically or foreign tariff-free. It was just up to them what they wanted to do. And there was a reason why there wasn't that much steel subject to the tariff. We had the USMCA, so Canadian steel and Mexican steel, our two largest steel trading partners, was not tariffed. At the time, we were in discussions with the EU about this thing called the global arrangement. They're going to try to tie climate change to tariffs and use tariffs to reduce emissions. That fell apart like a $5 watch. But nonetheless, that steel wasn't tariffed. Some big steel-producing countries, Brazil, Argentina, South Korea, they had quotas, so they weren't paying tariffs. As long as they stayed within their quota, they didn't have to pay tariffs. And then you had this exclusion and exemptions process that was very opaque and very confusing where people could game the system and file for a tariff exemption and get it. And that wasn't subject to tariffs. So he cleaned all that up. So because I remember when that first round of tariffs came out, was that a 25? What was that? 25 percent? Yes, 25 percent. And I remember when I was buying metal that was import from mills. you know let's say at that time i was buying it for you know 60 cents a pound landed and it was going to cost me a buck something to buy domestic The 25% still wasn't closing the gap. No. You know, it was still cheaper after all of that to still just import it. And then you're talking about our trading partners, Canada and Mexico, which became just a routing opportunity for bringing it in. Instead of bringing it directly to like Long Beach or directly into Houston or otherwise, you just say, well, that's fine. We'll take it into Canada. and cross over that border or we'll take it into mexico and cross it over that border absolutely and so it's your it's one of the big effects of this most recent iteration of this tariff uh law executive order is that it's closing off sort of the loopholes that were undermining what was the initial intent. No, you're absolutely right. It's called transshipment, where you take steel that should be tariffed, you send it somewhere else, you might even do a minor alteration to the steel, and then you get it in tariff-free. So yeah, that, circumvention, subsidies, state ownership, you name it, all of that was gaming the system. And you talked about earlier, I was listening as you were talking, As I should be. And you said that you've seen $30 billion of investment into kind of this post-tariff market, which is something that I've discussed with other people, which is when you create tariffs, you create voids in markets, right? Where, okay, we did have foreign steel coming in and filling that demand. And then you say, well, I love the idea of tariffs in free and fair trade, but it's just going to create sort of a lack of supply in places that need demand. So if I heard you correctly, the U.S. or domestic market has actually responded to filling what would have been demand from other steel sources. That's absolutely right, Shep. the members of the SMA have done is they just haven't rested on their laurels, okay? They're saying, okay, we have this unique period we're in now. Let's invest. Let's put some new capacity online so we can be even more competitive than we already are. A good example of that investment is in the state of West Virginia. I have two members, one, Nucor, who's made the largest capital investment in the state's history to build a steel mill there. It's huge. It's a game changer. But more importantly, Shep, it's a life changer. Think about all of the people that are going to get good-paying jobs in West Virginia with health care, with daycare on site, with a clinic there. That's fantastic. Commercial Metals Company made a huge investment in West Virginia as well, and there are investments going on throughout the United States. Now, there are areas where we still need to import steel. One of that is electrical steel. There is still a shortage of the steel that we need, you know, grain-oriented electrical steel and even non-grain-oriented electrical steel to build transformers for cars, etc. So we've got to get going there. The problem is that type of investment is one that's highly technical and you've got to have demand. to make that kind of decision. As far as creating our own facilities to make those sorts of things. Yes, exactly, to build a greenfield facility for that. And are those... mean, those are things that are being discussed currently and just sort of like, oh, we can't get there? No, they're being discussed and contemplated all the time. And keep in mind, a good steel company is going to make an investment not based on whether they have a tariff or not, but based on whether the market demand is there. Yes. And so that's a big issue there. Also, you've got to have some kind of support for those folks that are first movers, the risk takers. You know, don't say, hey, I want you to build me a $3 billion facility, but I'm not going to help you if it takes a while for the market to ramp up. Or if you can't get a permit approved because of some bureaucrat in D.C. not willing to sign it off. So we need some of that as well. And that's the kind of stuff we work on. You do need support, which is that's the other part of tariffs, which is, OK, we're going to reduce what is likely reduce the inflow of foreign steel. And that's fine. But like I was talking about, you're going to have, oh, now these manufacturers of. products don't have that raw material coming in anymore. And they're kind of in a lurch. So I was thinking, well, how do you solve for that? Because we, the U S can make, we can make raw materials, but it's going to require that, you know, Some of these projects just get stamped so they can get made like that. That's really was the limiting reagents and getting those voids filled. And so you're seeing sort of what is a reasonable amount of cooperation to get. projects started? Yeah, I think we're seeing the beginnings of it, but we still need serious permit reform. The longer you take to approve a project, the more likely it is market conditions will change or companies will just change their mind. They'll say, hey, I can invest this money some other way. So that's important. Also, we need other policies that just help steal demand. You need good tax policy. You've got to get interest rates down. You have to be doing things so that there is the underlying demand there to build these projects and build them all the way to fruition. The one thing I'm most proudest of, Shep, is members of the SMA are actually moving earth. We're not just issuing press releases. We're actually building these mills, and we haven't stopped, no matter what the political environment or economic environment was, whether it was Biden, whether it was Trump, whether it was low interest rates or high interest rates. All of these projects are full steam ahead. You can't tell me one project that an SMA member has canceled, but yet you go over and look at the EU. They're canceling stuff left and right because they have high energy costs and they don't have demand. And the interest rates. play in because, you know, people need to get financing to get deals, to get things built. And I was assuming that's what you were talking about. But is that the importance of the interest rates? Yeah, I think that's the importance of the interest rates. Also, lower interest rates puts more money in the pockets of business owners, job creators, and consumers. And all of those folks drive steel demand as well. And, you know, building these facilities, I'm assuming, are also joint ventures of, you know, these are huge capital outweighs. And what are, you know, the kind of the JV market like for this? Are people in the investment community for these types of projects, are they feeling unsure of, well, you know, the ground could move beneath your feet? And I would say something from the steel industry, I would say, well, here's one thing I know for certain is we will always need steel. Yes. You know, so the ground can move as much as it wants. Right. But if we build this well and we get this moving out into the market, I can tell you that we're still always going to need steel. Yeah, you're right, Shep. What sort of advancements when you talk about, OK, we go from 10 man hours for a ton of steel to 0.5 man hours to a ton of steel. What is happening that allows for that level of efficiency to occur? Because steel, I think in the minds of many, we were making broadswords out of steel. This has been around for a minute. Now there's the economy of scale, which is a big part of it. But you would think that when you see jumps from 10 to 0.5, That's not an asymptotic thing that's occurring. That is like... It's exponential, yeah. So where are those gained? Well, I think you see the gains in a couple of things. Over that 50-year period, when you go back to when it was 10 million hours per ton, we were predominantly an integrated steel industry, blast furnaces, basic oxygen furnaces. You saw the growth of EAF. which is inherently more productive. And that's electric arc. Electric arc furnaces. You also saw some kind of groundbreaking things like thin slab casting. continuous casting you saw the birth of something called the mini mill or the micro mill where you become specialized and you said you know what we can get really productive if we make this one thing and make it really well and just go straight from the furnace to casting to rolling so you saw uh improvements there you also saw improvements in a in furnace technology where you're looking at ways to optimize energy by determining what kind of mix you use with your scrap. Do you supplement it with an ore-based metallic? Do you try to find ways to use things that get what's called an exothermic reaction? I'm getting real granular here. Versus endothermic. Yeah, there you go. But that helps you with your energy. And then you do something on the human side, Shep, that I think really goes underestimated. You reward people for risk-taking, original ideas, and you pay them handsomely for that. You have good variable comp programs that make sure people are productive but also operating safely. The best ideas are not coming from some office. The best ideas are coming from some shop floor in Beaumont, Texas, or in Blytheville, Arkansas, or Corpus Christi. So you need to think about that. True. And something that I just said last week, when you talk about where did the great ideas come from, I bet if you did a statistic, just an analysis over time, that it's probably not from where it should have come from. Yeah. You know what I mean? It's because being well-funded is great, but it also comes with a bandwidth of thought. Meaning that you're going to have to, like, we've raised money to do this thing. Yeah. And we've got to stay between these guardrails because the person who's, you know, who's raised the capital or the organization has is, it's not so much they've said, this is how we're going to do it. But like, this is the prevailing thought. And so we've got to chase this down. Whereas the person on the shop floor and, you know, wherever is sitting there. doing this and being like, why are they doing it this way? This is so stupid. Yeah, don't they know? Yeah. Yeah. And so they know that that's a machine that's doing that. And they're like, the only way to affect change is, if it's that important to me, is I've got to create a groundswell myself. Right. And the companies that succeed, they listen to those folks. And they listen to what they're saying, they evaluate it, and they implement those things. And that's why they've been so remarkably successful. But also the reward aspect of it is good. You know, a typical steelmaker makes about $100,000 a year. And that's with some good benefits. Typically, the mill is one of the highest paying employers with the best benefits in the communities where they operate. And the one thing that makes me proud about being in this industry is you've never heard people say that steel mills don't pay a living wage or don't provide benefits to their workers. They provide great benefits, good compensation, and if you're willing to work hard, be a good teammate, and be safe, the sky's the limit for you. Yeah, being safe is a big part of it. And I will tell you, having traveled around to different countries and visited those steel mills, That's not on their top 10 list. I was at a steel mill abroad, and I did not see a railing. I didn't see a yellow line on the floor. I didn't see safety glasses or helmets. I was like, you know, this is all about the steel. Yeah, and that's what we're competing against, Shep, and that's why it's so important to have a level playing field. Yeah, and so you do have an inside sort of track on this. What has been the reaction from the non-domestic mills? Because this new version of the tariff is pretty comprehensive from a blocking perspective. So what's sort of been the, there's got to be some outcry that's happening. Well, of course there was outcry originally, but what we have seen gradually over time is that the EU, even parts of Latin America, even parts of Asia, are adopting policies that are very similar to ours. Because if you look at it, there's just one big bad wolf in this scenario, and it's China. China still, this day, makes over 50% of all the steel in the world. You know, the U.S. steel market on a good year is maybe 100 million metric tons. China makes billions. tons of steel and that still has to go somewhere well i was at a massive mill in china one time and i was walking through the warehouse of their completed steel that they've made. And it was, they made all kinds of metal. And they were laughing, literally, about how much the U.S. pays for their steel. They were laughing because they were walking through the warehouse like, that stack right there is for that country. That's for that country. And they told me specifically, they go, you guys, meaning you U.S. people, you say I want to grade this with that and that and that and that. They're like, it's all the same stuff. And they were laughing. I was like, well, all the same stuff, meaning that. uh you make it all the same and we don't you know we don't do the pmi all the time so we don't know or It is literally like you just make it to the highest grade and, you know, spec possible and you just stencil it and just charge more. But he was saying just like apples to apples, we pay like almost 100 percent more than other countries. Yeah. But chances are that guy you were talking to, there was probably a 50 percent chance that mill you were at was owned by the Chinese government. I think it was 100 percent. And so but he was saying, though, that. uh within that mill though yeah like those like he was walking by stacks uh that were you know this one was destined for that country this country in that country he's like same material and he's like you guys are paying x percent more for the exact same thing uh and i was just like why they're like because he goes the way you order it is he goes, you guys say you need this back, you know, with this yield, this tensile, da-da-da-da-da. And he goes, I mean, it's all the same stuff, though. And I don't know if it's true or not. It just caught me. It was like, ugh. To me, Shep, that's kind of scary. And I know you've got a procurement background, but you think about that if you're crossing the Bay Area Bridge out in California. And you got a guy that's kind of saying, well, it's all the same. And, yeah, it's kind of close to the spec. It's close enough. It's in the ballpark. Ish. Yeah. You don't want that to happen. And that's one of the things. They don't have the same environmental, safety, quality, and regulatory standards that we do. Now, some of them do. Some of these other producers don't. And I think that's a little bit cavalier on that person's part to say that, because critical infrastructure needs to be built to some very high tolerance. Yeah, well, in procurement, if you were going to buy Chinese steel at that time, you say you just buy 120% of what you need. Because you're going to toss 20%. Yeah. You know, and I said, well, I was saying, well, that's the 20% you noticed. Exactly. You know, so, no, your point is taken. And I don't think it really has to be just in critical applications. It's really about I need to get what I order. And I think that's a big part of it. But I'm glad to hear it because I don't, you know, what I do is downstream of what mills do. Right. But you play a very important role. I really like your B2B platform and what you're doing. Yeah, no, that's great, man. Well, I think it does a lot of things. And I think first and foremost, what it does is it streamlines the procurement process. Because that is quietly expensive. I don't mean just the steel itself or the metal. What I mean is the amount of time that it takes for a PO to occur. And if I want to take it even up to a national level, it's also about being competitive as far as getting fulfillment done and getting things moving. And what digital trade does, I think, is it also... chronicles the movement of metal, meaning that I'm not going to say blockchain or anything like that, but what it does, it creates automatically a provenance with it that helps, especially when you are talking more critical applications, where you need to chase down. Yeah, that chain of custody is codified. You know where it is. You know where it is, and that's one of those things that... maybe doesn't matter until it does. I think you're right about that. You're like, oh, business as usual, all is great. And then we got to find out, okay, what was the source material for this and all that. But bringing it back to what you guys are doing. So SMA, so give me kind of a nuts and bolts, a little bit of what is the interaction? of how you are literally working with the mills like what do you guys are you uh representing them in you know i don't even know if lobbying is the word i'm looking for but are you help them sort of uh promote what they need to get done like you guys believe in their causes and you help them get things done is is that Totally wrong? No, I think that is a very good explanation, and you can call it lobbying. And I'll give you a very recent example. I've got a very talented guy that works for me. His name is Brandon Ferris. He's my executive vice president. He handles our government affairs and our trade policy, and he recently, just a week ago, testified on behalf of our members about an impending tariff on pig iron. Okay. Okay. Now, I want you to think about this. Why would you put a 25% tariff on a raw material that's not available domestically? Okay. In fact, that would hurt the president's agenda to grow the steel industry. All of this investment that's being made, all of a sudden you're going to slap on the 25% on a raw material that is not available domestically. There's no domestic market for pig iron. And since the Russia-Ukraine war, Pig iron has been hard to find. We have to source a lot of it from Brazil. So this is a baseline input. Yeah, baseline input. Yeah, not a semi-finished or finished product. And so he testified on behalf of our members. He did something called a post-hearing brief, which was a very complex but well -articulated analysis of why we shouldn't have this tariff. And sure enough, we prevailed. And we do stuff like that every day. We work real hard on permitting reform. We work very, very hard on making sure that the 232s don't get watered down. So those are the types of things that we do for our members. And then that's just on the policy side. We also do a lot of other things. We collect safety statistics and share them broadly with our members, even our non-members. In addition to that, we have programming for rank-and-file people at the mill. We have committees on human resources. logistics and transportation, environment and sustainability, energy, safety, you name it. So it's kind of the optimal mix of policy and idea sharing and networking is a good way of describing it. Well, as we all go through our careers and learn more and more about how industries work, it is a battle to keep... You know, your industry, one that is fairly treated. Yeah. It's hard to do. It's not automatic. And it only takes one advocacy group for the need to be many others. Yeah. Right? Exactly. And sometimes those are folks that write bigger checks. Yeah. Or have 30 people fighting for their industry instead of 10. Should have brought a bigger checkbook. And people are going to look at that any way they want, but they shouldn't look at it as surprised. That's just the way things can go. But the flip side of that coin is, look, if your idea is really that great or what you're doing is that great, next time you should be able to maybe bring that checkbook with you. I think that... Great idea. You know, money follows ideas that are worth supporting. That's true. In general. Yeah. So let's let's move on a tiny bit, which is you said it well, which was the soup du jour. Yeah. And what is the soup du jour? Yeah. Right now, I think there's probably three or four big issues and we can just kind of take them randomly. One, we've talked enough about trade, I think. Let's talk a little bit about AI in the steel industry and in steelmaking. I think AI has enormous potential, Shep. It has potential to make steelmaking more efficient, more productive. Also, those ancillary areas that support steelmaking, sales and marketing, for example. Environmental, AI can do a lot to help there. But at the end of the day, there's another type of AI that I'm really big on in the steel industry, and that's called actual intelligence, not artificial intelligence. And when you make steel, there's still a human element to it. You've got to have good discernment. You have to have good judgment. You have to have a very strong fiber to make steel, and AI doesn't quite do that yet. But I think when you look at some of the applications that it can do well, for example, efficiency. There's a concept, I'm pretty sure you've heard of it. It's called creating a digital twin of your mill or of your service center, where you have this virtual version of your mill up on the cloud, and you can play around with what kind of material input you use, the residence time and furnaces, what kind of temperature you run it at. and see what might happen. That way you don't have to experiment in real time on the actual mill. Another area we see it is in predictive maintenance. You can use AI with its ability to crunch all of these numbers and look at all this data at one time. and figure out when a piece of equipment might start malfunctioning or maybe even break down based on all of these sensors and data inputs. You can also use it to geospatially map where incidents occur in your facility so you can know, hey, maybe that area needs a little more attention because it's not as safe because we've had umpteen slips, trips, and falls over there. Right. So I see that. It'll also help with customer service and procurement. But then again, Steel is always going to have a human element. Even though it's one of the most highly engineered products in the world, you still have to have human beings making it. And that's one thing that something based on a large language model hasn't quite shown me it can do. Well, I'm going to push back on a couple of things for the sake of discussion, not because I necessarily have any depth of knowledge in this. But just as a listener, Okay, so steel is the highest engineered materials. And I think that that's probably, I don't know, maybe not understood totally, but it's kind of like you think of it as like baking a cake. You can make them so many different ways for different things. And when you ask somebody what kind of cake they like, you might get different answers from everybody. You're going to get German chocolate from me, Shep. Okay. All right. There we go. but you like a specific kind of German talk. Like if I, if I put 10 of them in front of you, You might like one more than the other. Yeah, if it wasn't made by my mom or wife. Okay, see, now we're getting down. That's a heavy qualifier, you know? And I won't ask you if it's your mom or your wife the one we like more because that's a sticky spot. But what I'm getting at is I don't think it's too important. Let's just take it face value that this is a highly engineered, highly custom, like spec thing. But you're saying AI. It's not going to be going in and making this stuff. You're talking about the physical production of it. And then I look at, you know, I remember when we were growing up. I grew up kind of, you know, 80s, you know. I think my memory kind of came online in about 1983. You know, I was born before, you know, I was born in the 70s. But like my pictures of remembering things is probably early 80s. I remember, you know, you're watching the news and they got these robotic arms at the plant doing stuff. And you visit other mills around the world and you see quite a bit of automation. And I think, I don't know if that's really for the efficiency of money or actually keeping people out of harm's way. Maybe both. It's both, yeah. But talk to me more about the... you know, maybe your position it can't do, it can't make it as maybe more of a near, is that a near term position? Or do you feel like ultimately you got to have what the HI, the human intelligence is always going to be there? Yeah. Two kinds of AI, artificial and actual. Actual intelligence is always going to be there. And also, I think we need to make a distinction between automation and robotics and AI as well. And the automation and robotics are a very important part of our industry through the use of cameras, sensors. mechanical robotics. We have kept people out of harm's way. We've been able to see inside of the steelmaking process in order for that person with actual intelligence to make a judgment call. So I think that's important. I think when you look at AI replacing jobs, the steel industry is already so efficient. We've kind of done that already on our own. Don't forget, 0.5 man hours per ton. You can't get much better than that. But where the AI will do it, it's going to impact sales. It's going to impact your customer service. Look at what folks like you guys. That's where AI can be a game changer. But I think at that mill level. you're still going to need that mix of artificial intelligence, actual intelligence, automation, and robotics.