I highly recommend the book Accounting For Slavery to anyone even remotely interested in this topic.
You talked about the costs of buying a human being, which is already horrible. But once an enslaved person was owned outright, they were used as collateral in loans to expand operations.
This was incredibly common here in Washington DC. Think about construction. Building the Capitol required a ton of temporary labor, for example, so Congress contracted for both free and enslaved labor for its construction.
One of my favorite challenges to give data science candidates is to generate a dataset with a column for date and a column for attendance. With some added noise, attendance started at 1000 and decreased by two per day. I showed a year of data and asked them to predict the value a month from now. It was astonishing how many candidates come up with complicated approaches, like randomizing the data into train/test/validation groups (the wrong thing to do for a time series).
I don't think any of my applicants so far has double clicked the file to open it in Excel, plotted the data, and saw just how linear it was. You could follow the line with your finger and come up with a decent prediction. Maybe do that in ggplot2 or matplotlib if you want to show you can write code. Even a quick lm(dta$attendance ~ dta$date) would get you great results.
And the open source models are only months behind, so the big AI companies need to keep burning money on R&D with no end in sight. If OpenAI took a quarter off from model development, they might fall behind forever.
I agree. This is why I think Google has the long term advantage. They already have so much data. I can ask Gemini a question and it'll reference an email I sent a month ago.
I completely agree with you. Our family's first computer ran Windows 3.1. Moving to Windows 95 was a huge revelation about the potential that a computer could unleash.
If a company has a fundamentally good business model, I can just buy the stock and forget about it for years. I wish there were something equivalent for companies with bad business models. You can't just short something and come back ten years later to check the price.
I've been FASCINATED by Automats ever since I watched Bugs Bunny as a child. The idea that you could just walk up, look at what looked good, and buy it seemed indescribably awesome.
I cover thousands of miles a year on my feet. Darn Tough socks live up to the name. I've never put a single hole in them. I've torn up socks from most other brands in a few hundred miles.
I see two different things going on. First, there's some market for humanoid robots, and Hyundai is spending a small amount of its R&D budget exploring the options.
Second, I have to imagine that there are spillover effects for their other robots. Being able to make legs that are nimble is good for the ability to make grabber arms for industrial robots. So even if the humanoid product line goes nowhere, they could end up with better material handlers.
I think we'll start seeing humanoid robots doing boring things in controlled settings within a decade. Imagine that Hyundai built a robot that can empty a Hyundai dishwasher and take out the trash from a Hyundai trash can. I could imagine senior living communities and college dorms using them to go from room to room to keep things tidy.
There are tons of really impressive robots that can't relocate from where they're installed. At this point, they're so common that they don't get HN threads. I doubt Hyundai is betting the farm on humanoid robots, but it's a market worth exploring, and if nothing else they generate a lot of hype.
I spend about $10/year on scratchers since I like the causes they support. But most of the volume that I see being bought at the corner store down the street comes from folks who really don't appear to understand the odds.
Happy to see a shoutout to QBASIC in this thread. That was my first programming language, taught in 10th grade intro to programming. I loved it! I then took C++ at the local community college then AP Comp Sci in 12th grade.
My friends in that sector all seem to have a lot in common with any other job in a massive corporation. Lots of red tape, hard to advance, but you know your paycheck will clear the bank and nobody's demanding that you put in 100-hour weeks. The difference is the product you're working on, which can range from weapons systems to more mundane things... the "Defense industry" ranges from the assembly line making bullets to the janitorial companies specializing in maintaining military facilities.
In my experience, infosec jobs are a mix of being bored (as you describe), doing meetings and documentation (e.g. Sales has a prospect that wants assurances that you take security seriously before signing a contract), and absolute blood curdling panic when there's an event.
If your infosec folks are always working 95%, they won't have time for the second two categories.
I've worked in all three sectors. My experience in the federal government suggests that government jobs are clearly defined, but not necessarily easier than a corporate job. If you want to do a specific thing for the next decade, you're better of going gov. But if you want to make a lot of money and don't mind your job changing with every CEO transition, go corporate. And if you want somewhere in the middle, go non-profit.
You absolutely nailed it. I'm the Director of Advanced Analytics in a company called Cision. I've spent a lot of time analyzing content quantitatively. It's easy to find spurious correlations in this kind of data and present them confidently.
I have had a similar experience. I think one big problem is that EA often uses a low discount rate, meaning they treat theoretical people who won't be born for a century with similar value as people who are alive today. In theory that's defensible, but in reality it means you can hand wave at any large scale issue and come up with massive numbers of lives saved.
My church has a shower ministry, where we open up our showers to people without homes so they can clean up. We also provide clothes and personal supplies. That's just about the opposite of what EA would say we should do, but we can count exactly how many showers we provide and supplies we distribute and how those numbers are trending. Shouting "AI and asteroids!" is more EA, but it eventually devolves into the behavior you describe.
I wonder how aggressively this would be enforced. Like, if this website let people place an order a year in advance for a specific price, would the regulators come knocking?
Huge fan here! I read the article when it was new and was enamored with the story. I ordered a small box. The onions were terrific. So I order a large box every year.
Any political ideology could be bolstered with the heavy handed use of surveillance technology. That's why groups like the ACLU and EFF exist. They both vehemently agree that there's no such thing as a good police state. Even if your team is running it.
If you want specific examples, I'd point to the Freedom of Access to Clinic Entrances Act. The left wants to throw people in prison for years for protesting while the right wants to pretend it doesn't exist and allow women seeking abortions to be intimidated. I'd also point to the levels of technology deployed to catch the January 6 perpetrators. I live in DC and have a hard time doing a both-sides here since the experience of living in a city under attack is still rather raw; whatever your views, you can't describe the right's response as "enforcing the laws as written."
I really wonder if we're going to see a reversion to old school project management, with a PMBOK's worth of detailed project documents for every major initiative instead of the modern Agile/Scrum/Kanban approach that seems to work better for human devs. If you can get everyone to agree on the minutia of, for example, the Stakeholder Management Plan, up front then the LLM actually has a chance of developing a decent program that has the features everyone wants.
Then again, if humans could agree on all the project minutia at the outset, we never would have developed the other systems.