So I worked on robots for a fair bit of my career before quitting because the reality just didn't mesh with the hype. I've worked on self-driving cars, compute

So I worked on robots for a fair bit of my career before quitting because the reality just didn't mesh with the hype. I've worked on self-driving cars, computer vision applications for automated surveillance, physical robots for warehouse automation similar to what is being described in this article, and more. This is ignoring all the problems with these systems. Workplace injuries are completely ignored, and I have never worked somewhere with a physical robot that did not harm someone at some point, no matter how seriously safety was taken. The reality is, with current tech (which is always improving!) that robots are more dangerous - full stop. https://www.bbc.com/news/technology-54355803 https://www.bbc.com/news/technology-54355803 It talks about the simplicity of these systems which is true, but the problem is they're so simple something as weird as changing the color of the box can completely break the system with no resolution. I made up that example because I can't talk about real things that broke the workflow for warehouse automation companies due to NDAs, but they were equally stupid. Basically if you wanted to change anything then the robot usually had to be scrapped and redesigned, which costs millions and takes months or even years. It mentions the theoretical gains of using cheaper robots to replace expensive labor, and this is how these systems are sold to everyone who has never really worked with robots. Speaking for the robotics company, the upfront cost of the robot was usually more than they ever recouped from revenue. When you factor in the cost of maintenance (these robots are monitored by people that make a lot of money), not to mention the R&D then it never turned a profit. Speaking for the warehouses, they would frequently complain about the robots breaking, the inability to get work done, and the biggest complaint was "it's worse than the humans it's supposed to replace." The technology just isn't there yet. I don't know when it will be. If you are interested in research then robotics is a fascinating field. If you are trying to make money by solving real problems that aren't subsidized by VC then it's demoralizing as hell. Most robotics companies never make a good product. Amazon has Kiva Systems which work great, but you'd be surprised that most of their competitors still never figured out how to have positive margins on their products. iRobot had the Roomba which is still going strong with lots of competitors. There are a handful of companies that sell robotic arms that make money. There are a handful of companies that sell sensors that make money (a lot less than you'd think). There are contracts you can get with the military that usually go nowhere. Rodney Brooks, who cofounded iRobot, failed with his cobots approach at Rethink Robotics. There are lots of other failures I haven't mentioned as well. Robots have a long way to go before they're seriously competing with humans.