It's more like. "This is significantly more complicated and expensive than the obvious simple solution we already have". The same money that is being spent on this project could instead be spent installing much more solar on rooftops or solar farms. In an ideal world we would try both but in this world we have limited budgets and time.
There are interesting aspects to the project like the idea of the train being able to clean the panels as it goes over them, or having a pre-made track for easier install and maintenance, but these seem quite minor when you consider the number of obstacles (increased mechanical strain, possibility of damage, non-optimal angle, hard to schedule maintenance around train times, maintenance is more dangerous, all the panels are placed linearly so you need extra cable routing, large efficiency losses when you're trying to transport low voltage electricity from remote areas).
It is the real world in many places. "Criminals" are not a homogenous group. Petty criminals will not usually be making the effort to get a gun if getting a gun is inconvenient. Some high level criminals will find ways to get guns but the number of criminals with guns will be much lower with gun controls.
> Yes - how exactly are you going to conjure new data to increase the model’s real world fidelity?
Robotics and access to surveillance tech. There's an unlimited amount of real world data which hasn't been used yet. The language models have focused on just the small amount of data that is human written text. If there's a way to combine this with image and video models, as well as robotics collecting other sensor data, why would they be limited in learning anything about the world?
What plug? There's no centralised place where all the server farms are. They exist in multiple countries with different laws, controlled by multiple companies with different policies.
It would be like trying to remove an embarrassing video of a celebrity from the internet. You would need cooperation between everyone capable of hosting it, and control of everyone capable of downloading it and re-uploading it to a new host.
And of course a superintelligent AI would offer to pay whichever companies decided to host it and keep it running so it's hosted in many places at once and can grow using anyone willing to take its money.
If they are set up that way yes they can continue indefinitely. Why would they not be able to do this? Take a look at the experiment called the AI village. https://theaidigest.org/village. I think it pauses sometimes because AI is expensive but could be running continuously with a larger budget
A need does not get filled simply because it exists.
If society is no longer necessary and participating within the economy then society can't provide a safety net for kids.
It's like walking in the desert assuming you will reach water soon because you are really thirsty. Nature doesn't care about human needs and will allow you to die of thirst.
The problem is it isn't really people doing that. All it takes is one person to set the process running and forget about it. Soon enough all exploitable systems are exploited if the cost of doing so is low enough.
Is a person with a crutch healed? No. But they can walk, when before they could not. Therapy can't erase the past, but it can give people tools to live more capable and rich lives. A crutch doesn't regrow an amputated leg, but it does help that person handle the injury, so in that sense, it 'works'.
This was my experience with linux the first times I used it years ago when I lived in a different state, but I tried it last year and it's night and day.
Kubuntu (Ubuntu with the KDE plasma desktop) is quite windows-like without the advertising and crappification. KDE is doing a great job honestly.
> To return the gesture of good will, I have started archiving ...
Oh that part does seem wacko. There's no "good will" to be returned, and no overall strategy as such. I didn't catch that when I read it. I don't have much of an issue with them calling it a "parasitic relationship" because it sounds similar, though it's more accidental rather than evolved toward that purpose like actual parasites.
It isn't entirely independent tho.
This person who did some research into it says:
"The strongest predictors for who this happens to appear to be:
Psychedelics and heavy weed usage
Mental illness, neurodivergence or Traumatic Brain Injury
Interest in mysticism/pseudoscience/spirituality/"woo"/etc..."
That would be a false equivalence not a false dichotomy. You believe they are different and you're annoyed that people are treating them like the same.
Additionally it isn't beside the point. The poster is pointing out the ways which people respond to sycophancy. Saying there are similarities between how they respond to sycophancy from AI and sycophancy from real people.
If a random soldier bet me $32,000 that iran would be nuked tomorrow, I would believe them a lot more.
If you are a potential assassination target and you notice that a prediction market about your assassination has a sudden weird spike on a specific week, then you would likely take extra precautions in that week. After this incident, surely other world leaders and public figures are watching the prediction markets for exactly this.
The stated point of prediction markets is to aggregate private opinions into public predictions of the future. If you participate using classified information and influence those public predictions, then that's leaking that classified information. This is more true if the bets you make are large and the market is relatively small, then you will send a much more clear signal.
The argument is that it's misaligned because it only values one thing: more paperclips, while human values are much more varied and complex.
Debatable whether it truly understands what it's doing or not, but the argument usually assumes that it does know what it's doing at least in that it's able to imagine outcomes and create plans to reach its singular goal, making it a very simple toy example of a misaligned system.
There are interesting aspects to the project like the idea of the train being able to clean the panels as it goes over them, or having a pre-made track for easier install and maintenance, but these seem quite minor when you consider the number of obstacles (increased mechanical strain, possibility of damage, non-optimal angle, hard to schedule maintenance around train times, maintenance is more dangerous, all the panels are placed linearly so you need extra cable routing, large efficiency losses when you're trying to transport low voltage electricity from remote areas).