What is “Vertex”? A mod-only thing? I don’t have that.
Richard_Kennaway
Karma: 6,036
If (after giving it a reasonable go) you can’t tell whether what you are doing is working, it’s not working.
You are assuming that there is some “real” opinion you had, but have imperfect knowledge of, as if we all contain a perfect reasoner about probabilities but have only imperfect access to its conclusions. I see no reason to imagine this phantom. We gave a probability; assuming we were honest, if anything was our probability at the time, then that was it.
If we see something exceptionally surprising, we may legitimately in hindsight re-examine how we arrived at the low probability we had assigned to it. We might detect errors in our assessment back then of the evidence then available. That does not change the answer that we gave, only the probability that we now assign to the event given only the information we had then.
Discovering the right answers to a set of examination questions does not allow us to claim that we “really” knew the answers when sitting the exam.
The English idioms of “paying attention” and “giving attention” are misleading. One is not paying or giving these as things to the other person, as a consideration for which they will pay or give some quantity of theirs to you. Instead, both are jointly using their time and attention to create something to their common benefit.
I can imagine people treating their paying of attention in that transactional way, an hour for an hour, a pat on the back for a pat on the back, but it strikes me as dysfunctional.
So, if Trump was miraculously saved, is that a reason to vote for him? After all, the extent to which your vote matters is the extent that you risk getting Novikov’ed out of existence.
The things that are given in a friendship are things that when you give them, you still have them. This is unlike buying a loaf of bread, where I am little concerned to support the baker, nor he me.
What do you think of barefoot shoes?
That’s a possible position, but I’m not sure how one can hold it and at the same time edit Wikipedia in good faith. Does anyone know how that can be justified?
Excuse me while I channel Insanity Wolf:
GOOD FAITH
IS DOING THE RIGHT THINGTHE RIGHT THING
IS WHAT I DECIDE IT ISWOULD YOU FOLLOW A RULE
IF IT WAS WRONG?BREAK ALL RULES
WHEN DOING RIGHT REQUIRES ITDOING RIGHT ISN’T EVERYTHING
IT’S THE ONLY THINGONE’S OWN OPINION
IS THE ONLY OPINION THAT COUNTSDOING GOOD
JUSTIFIES EVERYTHINGIF YOU ARE GOOD
ANYONE DISAGREEING WITH YOU IS EVILEVIL
MUST BE DESTROYEDAnd so on. I have getting on for two thousand of these aphorisms, all of them inspired by things people have said, all of them insane.
I don’t want to aggravate the OP’s problem, but a reason that immediately occurs to me is that the people running the upload have a statutory duty to do so but don’t actually care. Consider the treatment of the elderly incapable in some care homes.
Well, that is your third assumption. But it requires that the multiple instances are identical: if one is dying, all are. If one is revived, all are.
To adapt Woody Allen, I don’t want to achieve immortality by imagining that someone else is me, I want to achieve immortality by not dying.
The same argument refutes quantum immortality, or at least removes its attractiveness. Suppose I am facing imminent death. For example, I have a terminal illness and within a few days I am sure to expire. What hypothetical being differs from me in a single piece of quantum uncertainty resolving one way rather than another? Someone with that terminal illness who in a few days will expire almost as certainly as I will. You have to go very far from that person on the death ward to reach someone who is hale and hearty, and what does that person have to do with me? The most similar person to me who is still alive when I die is one who is just about to die. And there I will be stuck, endlessly dying until some other outcome rises to greater probability than my increasingly unlikely survival, and will that outcome be any more welcome? That is what quantum immortality would look like. I believe this point has been made by others, that QI should be considered horrifying rather than an escape from death.
The closest we have to coexisting but identical persons in this world is identical twins. I have never heard of anyone going to sleep as one of a pair of twins and waking up as the other.
If “I” am dying, but the other supposed instances of “me” are not, then the similarity between us has already been lost. “I” am on my own and will die alone.
“ChatGPT is Bullshit”
Thus the title of a recent paper. It appeared three weeks ago, but I haven’t seen it mentioned on LW yet.
The abstract: “Recently, there has been considerable interest in large language models: machine learning systems which produce human-like text and dialogue. Applications of these systems have been plagued by persistent inaccuracies in their output; these are often called “AI hallucinations”. We argue that these falsehoods, and the overall activity of large language models, is better understood as bullshit in the sense explored by Frankfurt (On Bullshit, Princeton, 2005): the models are in an important way indifferent to the truth of their outputs. We distinguish two ways in which the models can be said to be bullshitters, and argue that they clearly meet at least one of these definitions. We further argue that describing AI misrepresentations as bullshit is both a more useful and more accurate way of predicting and discussing the behaviour of these systems.”
Did you put this paper anywhere? I didn’t find anything on arXiv meeting the description.
I suddenly realized something: there is no such thing as absolute truth. Instead, there is only the ability to convince others of our point of view. How could I ever think that there were right and wrong answers?
I can’t tell if you’re being ironical here. I hope you are, because this flies in the face of the entire project of LessWrong, and for that matter, all of science, and, well, pretty much everything that is not of the Devil[1].
- ↩︎
I am being metaphorical here. But St Aquinas would say the same and mean it literally.
- ↩︎
I cannot tell from this what you mean by the word “bias”. If I replaced it by an explicit hole like “flobzerg”, I would not be able to tell from the context what it could possibly be. That we have preferences and act on them? That different people have different preferences? That different actions are possible in the same circumstances? No, “bias” has to do with systematically faulty cognition, but I am not seeing that in your examples.
I believe the past tense is actually “shat”. Much more expressive.
There’s a crack that the speaker touches on, that I’d like to know more of. None of the people able to understand that the safety reports meant “FIX THIS NOW OR IT WILL FALL DOWN” had the authority to direct the money to fix it. I’m thinking moral mazes rather than cracks.
Applying this to AI safety, do any of the organisations racing towards AGI have anyone authorised to shut a project down on account of a safety concern?
Out of billions. Unfortunate for the millions, but the billions have it.