Abhas Gupta, who has worked with five AI companies, argues that fears of AI-driven job loss aren't just a repeat of the Luddite fallacy. While past tech disruptions created new jobs, he contends that AI automation is fundamentally different and society isn't ready for it.
Chapters
Thank you. Thanks so much. All right. So thanks for having me, everyone. Thanks for being here. I appreciate that. This talk is a little like— it's a lot more conceptual. So I'm sure you guys have had much more practical conversations. This is going to be a lot more conceptual, theoretical, and a little bit of— A little bit of a scare. I think that's sort of where I end it, is that shit's going down and we're not prepared for it. So that's sort of the— some political, and which I really like that you guys are from all over, because everyone has different sort of political leanings, and it's really, it's interesting for me to hear how people from different backgrounds kind of respond to this. So, all right, so let's start. The title of my talk is ML Automation and a Jobless Future. How many have already heard people talk about a jobless future? We just discussed it, actually.
Oh.
Tell me, do you mind giving me a quick overview?
Who was the team who did the AI stuff? It was Blair and Ernest. But we were kind of doing future projecting, and they were talking about how there was going to be— well, you can explain it.
Yeah, well, we sort of skimmed through a book that spoke about this. But it was the concept that artificial intelligence will take over labor in the service market, and what that's What was the book?
Michio Kaku's book, Physics of the Future.
Oh, OK, OK. Oh, interesting. Yeah, all right, well, this may be a similar talk. I think the framework you'll see here is unique. I haven't seen anyone use this framework yet, so I think this kind of explains it in a little more visual way, a lot more concrete way. So, all right, so let's get started. So, quickly on my background, aside from the Fun Seeker stuff, I work on 5 companies in particular. So, this is across the board. BuildDirect, an ecommerce company. Workfusion, an enterprise workflow automation company. That is some really cool shit. A lot of this thinking has been driven from the work that we've done at Workfusion. Neon, which is a computer vision company. Earnest, it's in the fintech space. a lot of student lending stuff, and then Clover, which is a pretty innovative healthcare company. So those are the 5 companies I work with. They're kind of across multiple sectors, as you can see. What's consistent about all of them is that they are all deep AI and analytics companies. Maybe less so AI, but certainly all of them are analytics. WorkFusion has some pretty exciting AI stuff. And then my background, I did my undergrad in CS and computational biology. Spent some time in a biotech company, spent some time in an MD-PhD program, decided that was not a fit, left, did some strategy consulting work, and then have been at— in the venture space, first at Mohr Davidoff for the last 5 years, and then now Wildcat Venture Partners for the last year. All right, so how many people are familiar with the Luddite fallacy? Okay, so let's start there. The Luddite fallacy So the Luddites were these textile workers in the early 1800s that basically had these jobs putting textiles together or sewing them together, and then these machines came to kind of automate the textile work. And so they totally freaked out. All these people were put out of work. They protested, and they kind of vandalized all these places, and their argument was that all these machines are going to basically put us all out of work. And it turned out that what these machines did is actually— certainly they took jobs away from the people in the near term, but those people were then able to go and get other jobs because it essentially increased productivity, which increased wages, which allowed more money in the economy, which created more jobs. And so there were always other jobs being created that people could go to afterwards. And over time, that kind of increased wages as well. So it became called the Luddite fallacy, that the Luddites, I guess, fallaciously believed that machines were going to take over all work. And so this is a term that you hear a lot. Here are 3 very kind of prominent folks. Marc Andreessen— this is probably a good time to say that I don't believe robots will eat all jobs. And specifically says, this sort of thinking is textbook Luddism. Steven Rattner, he's the auto czar under Obama's administration, who talked about a similar kind of topic. And then Alex Tabarrok said that, hey, machines are increasing productivity. Productivity has been the driver of wages for as long as we've been doing economics. And so people should stop freaking out, essentially. You have some very prominent thinkers who are saying, this is not a big deal, we've gotten out of it every single time. And I'll make the case here that this is a really big deal and that we're not doing anything about it. All right?
