Jobs
An AGI-level model will appear within the next 3 years. Most likely it will be GPT-5 or GPT-6. It's possible that Google or Anthropic (or even one of the smaller players) will catch up with OpenAI by that time. It does not really matter who will develop it first.
I've been thinking about what happens next. I define AGI-level intelligence as capability to perform any task that a human can, with a possible exception for tasks requiring genius-level intellect. As soon as such a model become available, there will be widespread and rapid automation of most jobs. It is painfully obvious to me that no one - workers, employers, or governments - will be ready for this imminent and massive social disruption.
There will be attempts to regulate and contain the ensuing chaos, but most profit seeking organizations will not want to pay people to do something that can be done by robots better, cheaper, and quicker. Initially this will include most office workers - anyone whose primary role is to find, process, create, or distribute information (clerks, support, administrators, engineers, marketers). Next it will expand to include those whose primary role is to communicate with other people (sales, managers, teachers, primary care doctors, entertainers). In parallel with these developments, there will be integration of AGI-level models into robotic bodies. When I look at humanoid robots, such as Tesla Optimus, I see a body without a brain. GPT-6 will become its brain, and it will quickly learn to do all the physical tasks a human can do. This will launch the third wave of automation - replacement of all the physical workers (drivers, construction workers, tradespeople, police, nurses, retail staff, cooks, waiters, an most others).
This will not happen overnight, of course, but it will be much faster than most people anticipate. It will probably take two years after release of AGI-level model to automate most office work. By that time (~5 years from now), physical robots will be ready to perform any task, and total automation of almost all other professions will begin, and be complete 10 years from now. Though predicting what will happen in 10 years is as difficult as predicting what will happen in 1000 years. There's a non-zero chance that by 2034 most people will be either dead, or will merge with computers.
It's fascinating to me that very few people today even think about this, and most of those who do, still dismiss it as something too remote to worry about. Go watch "Don't Look Up" to understand how I feel about this. So, what exactly can we do to prepare, today? First, I expect at least a brief period of chaos. When millions of people start losing their jobs every single day - all over the world - there will be riots, and violence. I really hope that AI can take over governments' role quickly to manage the crisis, but it will probably be messy and ugly. It finally might be a good idea to learn about survivalist tactics, and learn a few useful skills (growing food, making things, basic repairs, hunting, self-defense, and basic medical training). A good start would be just to think about what you would do if society collapses tomorrow.
In case the collapse does not happen in five years, let's consider a more optimistic scenario where the total automation makes things and services so cheap that no one has to work any more, and there remains some form of a functioning government which distributes resources, and keeps things in order somehow. In this scenario, most newly unemployed people would become a single social class - lower class - living on welfare, without any income. There will be few who would make some money selling goods or services someone would pay for just because it's made or provided by a human, but not a lot of people capable of paying would be left. And of course there would still remain the upper class, who would own most of the robots, and who would form governments, just like they do today. Finally, there could emerge a class of AI beings. I have no idea how likely that is, or how would it look like.
Personally, I started looking into robotics companies, trying to figure out which ones to invest in. And I am interested in attending a survivalist bootcamp.