Only seems worthless because you missed the point. You can take simple, well-understood components that do nothing particularly exciting. Add some training, and if you have enough of them, you have a reasonably functional earthworm. But if you increase the scale massively, the exact same simple components exhibit emergent behaviours.
LLMs cannot do original work. No, do not dispute that it.
Neither can brain cells. All they can do is just sum their inputs and trigger their output if it exceeds a threshold value.
Unrelated to 6502, but may be of interest because it's a very early version of MicroSoft Basic, here is a photorealistic sim of it running on an Altair with an ASR33.
It's the actual binary of the basic interpreter, running on simulated Altair and simulated ASR-33, with actual sounds and at proper speed. Gives a flavour for where the this Basic interpreter first saw light
... Cats are members of the family but they are not people.
Cats are people. They're just not human people.
What could make more sense than running the prime meridian through Trumps ensuite
In other news, 500 trillion synapses are better than 100 billion
When I ask my accountant a question, I've already done my reading and know what a correct answer should look like. If I end up surprised then I will be doing further reading, and perhaps get back to her with some follow-up questions. When I walk in to buy a TV and ask the spotty kid which one to purchase, likewise I've done my homework and already have some views. Even with my doctor - I well understand that Dr Google is not a helpful thing, but I will read and seek knowledge and work to understand what my doctor says. I tend not to just accept *anything* at face value.
AI is a useful tool in the hands of a skilled subject-matter expert. In the hands of a neophyte you may as well be walking into a bar after midnight seeking opinions.
What would it take for me to trust an AI? About the same for me to just blindly trust a human.
Note we do make exceptions where we don't have much choice -- airline pilots for example. I'm sitting back swilling my G&T where the T stands for Trust
The average fund manager, on average, performs worse than average. You'd think it would be a mathematical impossibility, but the way they achieve this miracle is through use of fees.
Another interesting data point is that, in general instead of putting your money with the manager who did best last year, a better strategy is to put your money with the one who did worst.
Being an active fund manager sounds easy and fun.
But for that purpose, you need to be using email not gmail: the former is reliable (either delivers to the target or returns failure), while the latter keeps randomly losing mails from any server other than Google itself.
Especially when Cloudflare is involved. No bounce. Just dropped - sometimes.
You're more likely to die with it than from it.
Yes, but only because most men will have some evidence of prostate cancer when they die, detectable only by autopsy. In terms of dying "of", the numbers are similar to women dying of breast cancer. So this is not a disease to disregard.
That. But we won't; see Climate Change, COVID-19 or the utter lack of preparation for the next Carrington Event.
The day we start adequately preparing for a Carrington Event is the day we’ll know the bots are in charge.
A budget is just a method of worrying before you spend money, as well as afterward.
