The pension funds are protected, but that does not mean anything close to your promised benefit is available for you to receive. It also creates a recurring liability that the company is on the hook for. This has historically worked out to cases where employees must either get a partial payout from available funds and tank the loss, or have the company fold and everyone gets laid off.
Pensions are a better deal until they catastrophically fail.
That's not unreasonable if you have good tests that hit the right elements. My experience about 10 years ago with the AP exams were very positive. The tests were good, and even though the classes were taught to the test, they were some of the best classes I ever took.
Building a ship containing delicate sensors that needs to get into space, arrive at the target, and beam back measurements on the first try with a very low tolerance for error is hard.
A quick google suggests to me that smoking trends are still falling in the US and globally on average. imo smoking is considered lame to americans as a sign that someone doesn't have their shit together.
What exactly is it? I get the sense it's basically a knowledge graph and an inference engine. But what was it actually doing in terms of submitting queries and spitting out takeaways in human readable form?
I'm extremely skeptical about the anecdotes about the game as an indicator of this thing's competence. It sounds unlikely that this thing actually encoded any sort of game state or nuanced simulations and was really just spitballing on vague strategies that just happened to find some cheese (twice?). I'm guessing they had to play the strats until one of them proved valuable, and it's kind of weird and surprising that they thought this was a good use of their time and model.
> during pre-training, there is never an incentive for the model to say "I don't know" because it would be penalized. the model is incentivized to make an educated guess
The guess can be "I don't know". The base LLM would generally only say I don't know if it "knew" that it didn't know, which is not going to be very common. The tuned LLM would be the level responsible for trying to equate a lack of understanding to saying "I don't know"
I remember playing one of these a decade or so after it came out. Found out there were hookers and a prompt on whether or not to have safe sex. Despite my attempts, I was unable to receive any penalties for ignoring the prompt.
My point is not that you missed on some sweet cash. My point is that it's an example of extreme privilege and wealth opportunities that most people cannot afford. The post comes off as anti-materialism and worldly and not caring about the money. But it's really the opposite. Your particular circumstances around wealth are what enable you to do this.
I find it irritating. And phrases like this one:
> I looked around at the people that had been the 10 or 20 years and it scared the shit out of me.
I really feel like epic games should be paying games to let people use their consistent character across games. It doesn't need to be significant. Just the metahuman mesh.
> 5. A negative view would be that AI coding assistants mean people don’t necessarily need “libraries” like this if an LLM can spit out the same results in a couple queries (rather than hunting around Github or the internet for pre-packaged code).
Why use a library when the LLM can spit of a unique reimplementation of the library logic every time you need it?