Episode Transcript
[00:00:06] Good morning, everyone. Today I want to talk about the UAW strike and the effect of unions and the issues with unions in the world of AI and electric cars. And I want to talk about what's been going on in AI and what the implications are for productivity and possible job loss. And a lot more of this is going to be discussed in the next couple of weeks as we launch our research on the post industrial economy. So the UAW strike is pretty shocking to me. They're asking for 40% pay increase over the next, I think, three years, ongoing health care benefits through retirement, full benefits for new workers regardless of tenure, and even a four day work week, which are benefits that no other workforce that I know of can even hope to achieve. And we also have to reflect on the fact that the three US. Auto companies are in a death march towards electric vehicles, competing with a trillion dollar company by the name of Tesla that is making a significant profit on each car. And the American automakers, the non Tesla automakers Tesla is American are losing money on all EVs. In fact, they're losing billions. So I would say the UAW's timing is terrible. And if I were the CEO of one of these car companies, I would be pretty disgusted with the situation that's going on. So let me give you my thoughts on unions and what I've learned. I am not a union expert, but of course I know a lot about HR and a lot about the workforce. The first is, of course, unions. By their nature, I mean, the reason unions exist is because employers, particularly in the industrial age, didn't really treat labor as an investment. They considered labor to be replaceable, and they didn't treat people well, and they figured if somebody didn't like their job, they could leave and they just replace them with someone else. So there were issues with safety, low pay, lack of benefits, lack of flexibility, difficult work, on and on and on. You know, the history. And so labor unions really played a role in the formation of the business environment we have today, where most companies understand that culture, employee experience, fair pay, equitable pay, good management opportunities, growth, careers are essential to running a good company. And by the way, the auto companies know this. I mean, at least the executives that I talked to there, the US. Automakers understand know Tesla to agree. Does too. I think Tesla's maybe not quite the same culture, but they're similar. I know that Rivian does because I know a lot of the execs there. So this is not lost on the automakers. Unfortunately, there's this legacy organization called the UAW that has inserted itself in the middle between management and labor. Frankly, the fact that they call their constituents labor is probably a problem for them because they're diminishing their value by considering them labor. One of the big issues we talk about a lot in our research is the distinction of management and labor is a really legacy concept, and we need to let it go. All employees add value. All employees are creative. All employees touch customers in certain ways. And that is becoming more and more true as we have more electric products and a more AI in the workforce. Now, the reason that this particular dispute is such bad timing is that you may know that electric vehicles require maybe 20% or 25% as many parts as an internal combustion engine. There's no fuel injection. There's no carburetor. There's no oil. There's no water pump. I mean, there's just a whole bunch of things that aren't there hoses. I mean, there's just a whole bunch of stuff. You don't need an electric car. So the assembly process, the number of steps can be radically simplified. A lot of these components can be manufactured in an integrated way. They're going to be need for fewer people. And so for the UAW to extort, in a sense, the big three on wages at this particular point in time is counter to the actual direction that all these companies are going. If I were the UAW, I would be negotiating something to do with future jobs in the world of electric vehicles, and I would be helping the automakers make this transformation as opposed to hurting them. And from what I've read about the current guy who's running the UAW, he's at war with these companies.
[00:04:49] So that's number one. Number two, of course, is what unions do, is they get in the middle between management and employees. They represent employees, sure, but they also stand in the middle. And so what tends to happen is there's an adversarial relationship, of course, but the employees actually don't end up with that much more. They might end up with less. And there have been studies, and I will put them in the links that I've been reading that show that labor relations may increase wages, but they tend to reduce employee satisfaction and employee feedback. There's a really interesting study that was done on a whole bunch of labor unions in Europe where they're much more sophisticated about this than they are in the United States. And they found that the thing that really turns on employees the most is having a voice in the company's future. And that has a massive impact on their engagement and their careers, and the union gets in the way of that. Union participation reduces employee satisfaction. In this particular study, and they looked at dozens and dozens of unions all over Europe. I don't think they looked at them in the US. And according to an article in the Atlantic, even today, either GM or Ford workers make $20 more per hour than Tesla workers. So Tesla is doing something that allows them to be happy enough to work there. And I know people who work at Tesla. They like their jobs it's a well run company. So begs to question what is going on with the UAW and whether it's helping the economy, whether it's helping the auto industry. I'm not sure who it's helping. I guess I suppose it's helping the workers. And the third thing, of course, is this political issue now going on where the Biden administration has decided to lean in. And I think what's happening here is actually simply raw politics where Michigan being a swing state, biden doesn't want to lose it, trump doesn't want to lose it. So they figure the best way to attract workers in Michigan is to come out in favor of the union, and Biden apparently is going to stand on the picket line, which just blows my mind, but that's obviously his decision. So there's some pretty bizarre to me dynamics going on know they will probably end up with a settlement. The wages will go up. The automakers will have higher costs. There's a really interesting article in the Wall Street Journal this morning that says that what's really going to get hurt by this is the supply chain. There will be fewer parts. It'll be harder to get your car repaired. I'll tell you, I am never going to buy a car from any of one of those three companies, ever. I mean, I'm just not my perception as a consumer is I would never do business with a company that runs this way or has this kind of relationship with its labor. I mean, I have to ask myself, the union worker who's angry and hates their CEO because their CEO makes too much money is putting my car together. I don't want that car. I don't know about you guys, I don't want anything to do with that car. Oh, and by the way, that's the other point I want to point out. Most of the arguments that the EIW makes about why this strike is needed has to do with the fact that their employers are making so much money. So they're really unhappy that the company they work for is really successful, making a lot of profit. The CEOs are making tens of millions of dollars a year, and therefore they want their fair share. Yeah, fair enough. Obviously people feel that way. A couple of thoughts there. First of all, CEO makes ten or $20 million a year. May sound like an awful lot of money, but that's actually not a lot. There's a lot of Silicon Valley CEOs that make five to ten times that. The second is we want our companies to be successful. If you're going to be a GM or Ford or Stalantis employee, I hope you like the company you're a part of. I hope you are loyal to that company, at least that you feel an affiliation with it. I hope you want it to make a profit. I hope you want its stock to go up. This kind of jawboning I don't think helps that and the way wages work in an economy like ours is it is a market of wages. Anybody who is underpaid has the opportunity to talk to their boss. Not, of course, if there's a union, but normally quit, look for another job, moonlight, get themselves educated and get promoted, look for a new job inside of the company. You have a lot of options. You don't have to sit there necessarily and wait for a union negotiator to improve your condition. And I think what this tells me is that in some sense, there must be a sense of powerlessness amongst these auto workers where they simply either psychologically or realistically feel that they have no power, no control, no authority, no autonomy, no agency in these jobs. And therefore they can't speak to management. They can't talk to the plant manager, they can't talk to the executives. And so they're going to turn over their negotiation to the union. And the union, by the way, collects dues, has $850,000,000 in the bank, and it has had, over the years, corruption problems of its own. So listen, I'm not against unions at all. I'm not against any form of collective bargaining. I know it's been very big for teachers. It's very big in the federal government. There's all sorts of reasons why it exists. But this particular one just really strikes me as bad timing and poorly positioned for the economic initiatives we're trying to do in this country. Now, the funny thing about it, the US. Government has funded almost half a trillion dollars towards electric vehicles, charging stations, infrastructure, tax credits, all sorts of things. Yet at the same time, our administration is also sponsoring a union move that's going to raise the cost of electric vehicles for our manufacturers. By the way, it's a global market. There's some pretty good electric car companies in Germany and also in China. So we're going to end up in one of these competitive situations again, where, just like Toyota came in in the 1970s and wiped out everybody with the small cars, we may find a lot of Japanese or Chinese electric vehicles running around. And nobody wants to buy the Ford 150 Lightning because it's too expensive or they can't get parts. And we don't want that to happen. So I got to kind of wonder whether this is a lot of politics, but economically it really strikes me as OD. So anyway, that's something I want you guys to think about, and I'm always interested in your comments. Second thing I want to talk about this morning is AI. We have done a lot of work on this. I've talked to a lot of companies, a lot of vendors. We have a whole article I published on hackathons going on inside of companies.
[00:11:39] And let me give you some thoughts. There's a mad rush in HR technology vendors to announce generative AI solutions. It's actually kind of dysfunctional because a lot of the vendors are doing exactly the same thing. They're creating an API into chat GPT. They're launching a job description generator or some other tool generator to generate content and they're embedding it into their application. While that is certainly a productivity tool, it's a nice thing to have. It's pretty much a commodity. I don't know how much better it is to do that versus to cut and paste it into Chat GPT yourself. So what's going to have to happen is all of these vendors are going to have to at least the ones building those kinds of tools. Listen, the first and second generation backend AI systems are very different. But these front end ones and I had a very interesting meeting with Textio this week and they showed us the new tool they're building or launching to evaluate performance feedback and performance conversations. And if you know Textio, they've been around close to ten years. They basically were doing generative AI a long time ago, didn't call it that. And what they do is when you type a word or a paragraph or a phrase into Text AO, it looks at that combination of words. In the case of a job posting for recruiting, or a performance management feedback or some kind of feedback, it looks at tens to hundreds of thousands of job posts that have similar words. And it gives you feedback on whether these words are biased. They might be discriminatory, they might be misleading, they might be off putting to certain people, to women, to men, to Asians, to blacks, African Americans. And it's amazingly interesting and useful. And what it is, is highly intelligent generative AI, the new product which helps you with performance management, does the same thing for performance appraisals. But performance appraisals are really feedback which we use all day when we talk to people. So it gives you feedback on how to be specific and actionable and positive in your feedback, but be clear in your feedback. And it's really, really interesting. And the reason I bring it up is, first of all, I think Textio is a really interesting example of an advanced AI company. But secondly, it teaches us a lesson that these off the shelf generative AI tools are going to be of somewhat limited value for some HR applications. If you try to crowdsource or generate performance management feedback, job descriptions, you're going to get a generic answer and you're going to want to tune it. Now, some of the vendors seek out, for example, and others have tuning knobs and they send prompts to the AI on your behalf. So you get a more curated, generative answer. But your company, your company's values, your company's culture, your company's mission, your initiatives, those things need to be embedded into these documents and generated assets that are created by these systems. And so this generative AI is going to need some help. And I think you're going to be sort of shocked to see at the HR tech show in the next couple of weeks. How many vendors have done this? I think they're all trying. They're coming down the path in different levels of speed, and some of them are very sophisticated in their thinking and some of them are not. One of the things I'm going to do at the HR tech conference, both in the US. And Europe, is I'm going to show you who I believe the trailblazers are. I'm going to highlight five to ten vendors that I think are way ahead of the curve. And I'm not telling you to buy things from them, but I think they're exemplifying where AI is going. Now that we're talking about it, let me talk about the productivity and costs of AI. There's no question that AI is going to generate a lot of productivity. There's a really interesting article I found a couple of years old from some guys at MIT about the organizational impacts of AI. Because once you get a more intelligent system inside of the company, it doesn't only improve individual productivity, it makes the organization smarter, because you start to learn things about your company that you didn't know before. So I'll put that in the links. But there's a cost to this, and let me just briefly mention this. We're building our AI system, and it's been really fun. We have a lot of people playing with it, and it's quite astounding, and we're learning what's under the covers. First of all, there's a cost of computing. These things are not free. If you use GBT Four, it's quite expensive. I don't know what Amazon and Google are going to charge for their cloud services, but the generative AI burn cycle, so you're going to pay for that. And if the system's not well optimized, it could be very, very high. And so you have to budget for that. The second is you have to develop and integrate and find the information that you want to feed it. And that information is sprinkled all over your company. It's repetitive, some of it's incorrect, some of it's out of date. That's actually a project. That's not a small project to do data management to feed it. And you have to create metadata around it so that the AI can find it appropriately. When you click on a link in Bard or a link in Bing, that shows you the source of a document, the only reason that works is because they were very careful about the metadata management when they loaded that data into the system. That doesn't come out of the box. Generative AI doesn't automatically identify where content came from. The neural net doesn't do that out of the box. Somebody has to add that feature in. There have to be security rules and privileges in these systems so that everybody doesn't see every piece of data. Because of course, I don't get to see the CEO's salary if I'm a mid level manager. I only want to see my team. So there's things like that and then there's user interface stuff. These things need a really useful user interface. I've seen a lot of them and they look pretty easy to use at first blush. But if you're using them a lot, they're not. So we need to make them more creative in the user interface and that's going to take some effort. So there's effort involved here. And then of course, as my old friends at IBM used to say, new software systems are like babies. They throw up and sometimes need diapers. In other words, they don't always do the right thing. So a new AI system is going to be something you're going to have to train, you're going to have to improve it, you're going to have to monitor it, you're going to have to constantly watch it. And I don't think that's a difficult or necessarily impossible thing. But this isn't black magic that's suddenly going to just revolutionize your company and you're just going to turn it on and walk away. So relative to productivity, yes, some jobs are going to be much less important.
[00:18:40] The person that's filling out forms in your company hopefully is gone already, but that person's going to go away. The person that's asking silly questions may be replaced by a chat bot. But those people and by the know, this is why there's a writer strike. Getting back to the unions in La. They're worried those people will have to find other jobs and training and maintaining and curating and analyzing these AI systems is going to be a lot of work. So there's going to be plenty of work there. I'm not going to make a comment on the intellectual property issues, but there obviously has to be some work done on open generative AI systems reusing content that was authored by others. And I know there will be legal work and lots and lots of discussions about that. So anyway, this is some stuff to think about this weekend. I'm not anti union. I'm not anti labor, of course. I just, from our perspective see this evolution to the post industrial thinking so urgent right now. We're going to have a labor shortage. There's going to be radical change in the types of jobs people do. Companies are pouring money into employee development. I don't know if the labor union intervention is needed or even positive in this particular point in time. We'll see how this plays out. Hopefully the big three come out of it in good shape and I hope all the workers in Michigan and other places that are part of this also feel that it works out well for them. And I welcome your feedback and comments. Have a great weekend.
