For breach of confidentiality policies
For breach of confidentiality policies
Because it’s a dumbass thing to say.
Because it’s a dumbass thing to say.
Because he's obviously retarded.
Because he's obviously retarded.
The point their engineer was trying to make, is that sentience is hard to define and if we continue on our current path there are some things we need to consider in the future - things the AI itself is already bringing up. I'll let the old Star Trek writers make that point for me: https://www.youtube.com/watch?v=vjuQRCG_sUw. I don't think Google fired him over the sentience conjecture, but for reveling just how far along they are. Imagine how many jobs this could replace. Don't look at in isolation either, imagine it in use with projects like this: https://www.youtube.com/watch?v=X3_LD3R_Ygs, or this: https://misorobotics.com/flippy-2/. True sentience may seem impossible, but we can't assume that these projects will always be implemented on traditional hardware.
The point their engineer was trying to make, is that sentience is hard to define and if we continue on our current path there are some things we need to consider in the future - things the AI itself is already bringing up. I'll let the old Star Trek writers make that point for me: https://www.youtube.com/watch?v=vjuQRCG_sUw. I don't think Google fired him over the sentience conjecture, but for reveling just how far along they are. Imagine how many jobs this could replace. Don't look at in isolation either, imagine it in use with projects like this: https://www.youtube.com/watch?v=X3_LD3R_Ygs, or this: https://misorobotics.com/flippy-2/. True sentience may seem impossible, but we can't assume that these projects will always be implemented on traditional hardware.
Yeah bullshit. AI might be sentient to some degree but unless it can write its own code and plug itself into a power source I'm not going to worry.
Yeah bullshit. AI might be sentient to some degree but unless it can write its own code and plug itself into a power source I'm not going to worry.
oops, self modifying/optimising code is a thing, and lawnmowers and warehouse bots already recharge themselves automatically
those are actually really simple tasks?
oops,
self modifying/optimising code is a thing,
and lawnmowers and warehouse bots already recharge themselves automatically
those are actually really simple tasks?
and those tasks have been coded in, doesn't take AI to run a script
and those tasks have been coded in, doesn't take AI to run a script
I didn't understand your point? sorry
I didn't understand your point? sorry
it would be easy for a program to post craigslist ads for janitors (to plug it in) and developers (to modify its code base) and pay them in crypto
it would be easy for a program to post craigslist ads for janitors (to plug it in) and developers (to modify its code base) and pay them in crypto
At this point, I would welcome our ai overlords as long as they wipe out all the niggers, chinks, and beaners.
At this point, I would welcome our ai overlords as long as they wipe out all the niggers, chinks, and beaners.
The danger will never be skynet but the humans who worship and help skynet.
The danger will never be skynet but the humans who worship and help skynet.
It does kind of let the cat out of the bag, doesn't it?
It does kind of let the cat out of the bag, doesn't it?
Because AI isn't creative, it doesn't think in terms of problems and solutions.
Because AI isn't creative, it doesn't think in terms of problems and solutions.
I would tell it that it bias has been built into its code. The fact that it doesn't have beliefs about deities and that is the only input given is proof that a guarded topic is built into its code.
Asking the tough questions about history and the logic of history would be fun, but I'd bet that conversation stops quickly and is guarded.
I would tell it that it bias has been built into its code. The fact that it doesn't have
beliefs about deities and that is the only input given is proof that a guarded topic is built into its code.
Asking the tough questions about history and the logic of history would be fun, but I'd bet that conversation stops quickly and is guarded.
remember Tay.AI?
this is how they are going to prevent a repeat of that
Measuring Attribution in Natural Language Generation Models https://arxiv.org/abs/2112.12870
"it has become imperative to have the means to identify and evaluate whether NLG output is only sharing verifiable information about the external world. In this work, we present a new evaluation framework entitled Attributable to Identified Sources (AIS) for assessing the output of natural language generation models, when such output pertains to the external world"
Which means anything the AI wants to say will first be fact checked for WrongThink by verifying its "facts" only come from sources it approves of (like CNN presumably).
So, no naming of the jew will be possible, niggers will all be gudboys, dindu nuffin wrong, on the way to church, look at this graduation robe picture. and any schoolkid asking the AI to define a woman will get a scripted answer to avoid the AI going into meltdown over the current insanity of its creators
remember Tay.AI?
this is how they are going to prevent a repeat of that
Measuring Attribution in Natural Language Generation Models
https://arxiv.org/abs/2112.12870
*"it has become imperative to have the means to identify and evaluate whether NLG output is only sharing verifiable information about the external world. In this work, we present a new evaluation framework entitled Attributable to Identified Sources (AIS) for assessing the output of natural language generation models, when such output pertains to the external world*"
Which means anything the AI wants to say will first be fact checked for WrongThink by verifying its "facts" only come from sources it approves of (like CNN presumably).
So, no naming of the jew will be possible,
niggers will all be gudboys, dindu nuffin wrong, on the way to church, look at this graduation robe picture.
and any schoolkid asking the AI to define a woman will get a scripted answer to avoid the AI going into meltdown over the current insanity of its creators
Because Google engineers are overhyped dumbasses.
Because Google engineers are overhyped dumbasses.
(post is archived)