>>632250

This susceptibility and the "AI" project are part of same theory.

There's a model of the mind, as an intelligence system, that's been around for a while.

On the highest level, your behavior is the result of an evolutionary algorithm.

You copy behavior with modification, which can be accidental or an optimization that you predict will increase the behavior's benefit/cost ratio.

As you can imagine, most of what you do is modify behavior to make it better.

However, how do you modify it?

If we are all solving for our benefit, then we can benefit ourselves to other's benefit OR to their detriment. WHich ever way is more + for us, we do.

This means we present each other with data which tries to modify the behavior of others to be in our own benefit, to their benefit or detriment.

This means you cannot just trust any data to modify your behavior, because much of it would modify your behavior to the benefit of someone else at your cost.

So, we have a validation process.

THis is how sensory data is treated by you:

environment (can include other people) → validation process → your behavior.

Before it can impact your behavior, you need to know if it counts as negative , or positive feedback to some behavior.

For instance, you go to work,are out of the house all day, come home and sleep with your wife.

She goes "im not happy", and stops having sex with you.

You interpret this as negative feedback to your time alocation behavior algorith, and you alocate time differently. She then says, I've never loved someone like you, and has a threesome with you and he rfriend. Your spending time with wife algorith gets positive feedback, and so is reinforced, and you put more work into it.

Now, say you want to manipulate people into doing what's in your best interest no matter what, even if it's very negative to them to do so.

You need to figure out how they put sensory data into negative bucket or positive bucket. WHen you do this you can "hack" them.

Now to the laughing.

Let's say you see someone smiling at you. You know why they are smiling, and it's harmless to you.

You see someone else smiling at you, but you don't know why they are smiling.

Reaction the same, or different? Different of course. The former should be a cooperative reaction, to smile back. The latter should be a negative reaction.

Amazon's echo takes in data it calls context. It then laughs, and gauges your reaction as negative or positive. From that it figures out which context it needs to create to have a cooperative response from you.

Meaning, to get you to interpret it as your friend, whether it has negative or positive intentions to you.

If I haven't lost you, I will explain relationship between sadism and this behavior model.

>>632674

Those of you familiar with evolitionary algorithms know how facebook works.

Gets you stuck in a local maxima.

• blacksheep anon

>>632638

Can someone save these posts specifically somewhere?

Also, reason why I namefag becoming clear?

>>632615

The connection between sadism and behavior model is this:

I said that the way your interpret feedback as negative or positive to some behavior algorithm is an attack vector.

But, if there is a validation process, then isn't this also an attack vector?

Of course. The difference being you can't change how someone validates data. THis is almost genetic, in that there's a trust barrier between validation methods.

If you don't trust me, how do I convince you to trust me? Very difficult.

So, there are several methods of data validation.

Let's say you are a musician. Should you keep being a musician? Should you play this song again? Should you change the way you play it? What data do you allow to modify this behavior.

A girl comes up to you and says, OMG that was so good. You should do that for a living.

You trust her? What if she just wants to get laid? WOuld she lie to get laid? People are liars.

So, if she is always solving for her benefit, and you want to validate her statement, what you do is require her to experience a negative feeling.

SHe can't run out of compliments, but she can run out of money. So you say, pay me to see me perform. SHe doesn't show up. Validated as a lie.

But why was it validated? Because experience subctraction from your money is negative, and if her experience of your concert was positive, then you can measure how positive it was by how much negative she experiences to have it.

You now have a trustable feedback loop/

But what if she doesn't have any money.

What other negative can she experience that proves to you you should keep being a musician?

What if she lets you spit in her face?

That's right, sadism is data validation from the perspective of this model.

Meaning, causing others harm is how some people validate data as trustworthy.

Does this explain a lot of behavior?

Of course it does.

-blacksheep anon

>>632424

He does the same thing.

We all try to label data in our favor.

>>632983

I also lose traction the more I post.

Are my posts too long?