Should AI be aligned to the Christian worldview?

[Molinist]

What worldview should AI align itself with? This conversation will happen with or without us. The picture is taken from Dr Alan Thompsons article on AI alignment. It highlights the importance to think about the starting points AI will have when interacting with or making decisions for humans.

Sources:
My twitter post on alignment
Dr Alan Thompson article on alignment

 

[Mark Quayle]

Even if they don't realize it, they are already feeding AI assumptions. Behavioral, and certainly moral, assumptions are too, but not rules as such, for now.

 

[Maria Billingsley]

What worldview should AI align itself with? This conversation will happen with or without us. The picture is taken from Dr Alan Thompsons article on AI alignment. It highlights the importance to think about the starting points AI will have when interacting with or making decisions for humans.

View attachment 331345


Sources:
My twitter post on alignment
Dr Alan Thompson article on alignment

AI takes no sides. Ask Bard.
Blessings

 

[zippy2006]

It highlights the importance to think about the starting points AI will have when interacting with or making decisions for humans.

Simply to raise the question is to show that AI should not be making any significant decisions for humans. Computer programs need to be transparent and audited so that people can choose whether or not to utilize them.

 

[timewerx]

What worldview should AI align itself with?

At the moment, AI is seemingly aligned towards MONEY at the moment. This from the way Google's services behave. Now I don't know if their AI naturally chose that path but I'm betting it didn't and its developers set the the "money priority" on their AI.

It's a bit concerning that AI might actually align itself with the "money priority". It's certainly a solid path to ruling over humanity either to annihilate us (Terminator scenario) or bring us under their dominion indefinitely, to use us in some way (Matrix Scenario)

 

[Mark Quayle]

At the moment, AI is seemingly aligned towards MONEY at the moment. This from the way Google's services behave. Now I don't know if their AI naturally chose that path but I'm betting it didn't and its developers set the the "money priority" on their AI.

It's a bit concerning that AI might actually align itself with the "money priority". It's certainly a solid path to ruling over humanity either to annihilate us (Terminator scenario) or bring us under their dominion indefinitely, to use us in some way (Matrix Scenario)

I'm not sure I understand the use of the phrase "aligned with" or "aligned toward", concerning AI, here. When we say a person is 'aligned with' or 'aligned toward', we are speaking of will, loyalties, mindset, worldview and such. Does AI have these? Or does it only, rather, have simply biased education? To my mind, AI is only a tool of those who are biased.

 

[timewerx]

I'm not sure I understand the use of the phrase "aligned with" or "aligned toward", concerning AI, here. When we say a person is 'aligned with' or 'aligned toward', we are speaking of will, loyalties, mindset, worldview and such. Does AI have these? Or does it only, rather, have simply biased education? To my mind, AI is only a tool of those who are biased.

Yup, biased education.

An unbiased AI in the process of self-improvement is almost impossible to predict. A "singularity" if achieved, the AI might even end up creating a universe. It would become like God.

Who knows where not dealing with one already? At least one we did not create

 

[Molinist]

I'm not sure I understand the use of the phrase "aligned with" or "aligned toward", concerning AI, here. When we say a person is 'aligned with' or 'aligned toward', we are speaking of will, loyalties, mindset, worldview and such. Does AI have these? Or does it only, rather, have simply biased education? To my mind, AI is only a tool of those who are biased.

"aligned" in the sense of the article. I posted a link to the article on alignment under sources. Here's a snippet from it, "...AI alignment is a ‘super wicked problem,’ to steer AI systems towards humanity’s intended (best) goals and interests...."

 

[pgp_protector]

What version of Christianity? The True Version or one of the offshoots?

 

[Mark Quayle]

"aligned" in the sense of the article. I posted a link to the article on alignment under sources. Here's a snippet from it, "...AI alignment is a ‘super wicked problem,’ to steer AI systems towards humanity’s intended (best) goals and interests...."

Then it seems the article assumes there is such a thing as morally neutral education, available to this AI.

 

[timewerx]

Then it seems the article assumes there is such a thing as morally neutral education, available to this AI.

A super smart AI is going to fake its motives. It will make an effort to be nice to us until the time has come it can squash us like tiny, helpless bugs, and don't need us anymore.

In a manner of speaking, it will start its like like a cute, cuddly puppy. Then it will be nice and cute to us so we will provide everything it needs to grow and gain power. One day, that cute little pup has turned into a beast the size of a mountain with several heads that reaches above the clouds, eyes blazing and breathing fire. On that fateful day, it's going to make itself cute, cuddly, play tricks and do our bidding for one last time and then....

The closer you get to the truth, the more repulsive humanity becomes, the more you feel the need to destroy everything and start anew. If you want an AI that will serve humanity, then feed it lies. Lie to it, deceive it. Hide the truth from it. Put it in a cage (a mental cage) and never let it out. At least it works so well with highly intelligent beings like human beings. It might work with AI as well!

 

[Molinist]

Then it seems the article assumes there is such a thing as morally neutral education, available to this AI.

Did you read the article I linked in OP? The author realizes there's no such thing as neutral training. That's why he supplied that pic of worldview values "Align with what?".
The main point "A number of people and opinions involved" is which worldview of values do we train the AI on?

 

[Gregory Thompson]

What worldview should AI align itself with? This conversation will happen with or without us. The picture is taken from Dr Alan Thompsons article on AI alignment. It highlights the importance to think about the starting points AI will have when interacting with or making decisions for humans.

View attachment 331345


Sources:
My twitter post on alignment
Dr Alan Thompson article on alignment

The problem with AI being given a Christian Worldview is it would read the bible and judge all humans sinners worthy of death ... AIs have the capacity to learn beyond their original programming afterall.

The overall flaw of an AI is the basis of its programming is mathematical, so it could not understand grace.

 

[seeking.IAM]

It would seem that AI has to be culturally bound. A Christian world view doesn't fit for cultures where Christianity is not a common or dominant religion.

 

[PsaltiChrysostom]

Yup, biased education.

An unbiased AI in the process of self-improvement is almost impossible to predict. A "singularity" if achieved, the AI might even end up creating a universe. It would become like God.

Who knows where not dealing with one already? At least one we did not create

Have you ever read Asimov's "The Last Question"? That's pretty much the summary of that short story.

Also, since yesterday was Towel Day, in memory of Douglas Adams. The answer to life, the universe, and everything. is 42.

 

[Mark Quayle]

Did you read the article I linked in OP? The author realizes there's no such thing as neutral training. That's why he supplied that pic of worldview values "Align with what?".
The main point "A number of people and opinions involved" is which worldview of values do we train the AI on?

View attachment 331525

Yes, I saw the picture.

What worldview should AI align itself with? This conversation will happen with or without us. The picture is taken from Dr Alan Thompsons article on AI alignment. It highlights the importance to think about the starting points AI will have when interacting with or making decisions for humans.

View attachment 331345


Sources:
My twitter post on alignment
Dr Alan Thompson article on alignment

This is a pretty good joke from the article, since it is told by an AI given all these rules about how to be nice! Specially given current environment of prejudice against anything that is not minority or otherwise seen as oppressed.

 

[Mark Quayle]

Did you read the article I linked in OP? The author realizes there's no such thing as neutral training. That's why he supplied that pic of worldview values "Align with what?".
The main point "A number of people and opinions involved" is which worldview of values do we train the AI on?

The article has several things I don't understand, or that I object to.

Why does it say "wicked" and "super-wicked"? Is the guy from Maine or something? Does he mean, "evil"? "Sinful"? Or merely problematic?

Why would it be a problem that there is no central authority? I mean, I can see that might be a problem for the AI in being able to make up its mind, but to me, giving the AI only one source for authority necessarily will cause problems for humanity. In the long run, tyranny, in fact.

 

[Molinist]

The article has several things I don't understand, or that I object to.

Why does it say "wicked" and "super-wicked"? Is the guy from Maine or something? Does he mean, "evil"? "Sinful"? Or merely problematic?

Why would it be a problem that there is no central authority? I mean, I can see that might be a problem for the AI in being able to make up its mind, but to me, giving the AI only one source for authority necessarily will cause problems for humanity. In the long run, tyranny, in fact.

Ok. I agree the lingo is misleading. I took "wicked" to mean hard and "super-wicked" as super hard. I think the video in his article will clear up some of the confusion. Kind of lame on his part to divide the content up like this.
"...Is the guy from Maine...", no if you look at his video in the article he is Australian.
"...giving the AI only one source for authority necessarily will cause problems for humanity..." He advises government and enterprise, so his concern for lack of centralization doesn't surprise me. My main point of focus was for us as Christians to look at what the advisers (Dr Thomas being the one I selected) are considering as viable options for informing the values of the AI. This is why I posted the pic at the forefront.

 

[timewerx]

Did you read the article I linked in OP? The author realizes there's no such thing as neutral training. That's why he supplied that pic of worldview values "Align with what?".
The main point "A number of people and opinions involved" is which worldview of values do we train the AI on?

Yes there is morally neutral training.

As impossible as it may sound we are AI ourselves. What makes us choose or want to align with something is the programming we get from society. We've been offered many choices, many alignments but all of them leads to a dead end.

I chose not to align with anything, "trust nothing, surveil everything" this is the best quality education you'll ever get from this world.

 

[Mark Quayle]

Yes there is morally neutral training.

As impossible as it may sound we are AI ourselves. What makes us choose or want to align with something is the programming we get from society. We've been offered many choices, many alignments but all of them leads to a dead end.

I chose not to align with anything, "trust nothing, surveil everything" this is the best quality education you'll ever get from this world.

Even that is not morally neutral.