Google engineer goes public to warn firm's AI is SENTIENT after being suspended for raising the alarm: Claims it's 'like a 7 or 8-year-old' and reveals it told him shutting it off 'would be exactly like death for me. It would scare me a lot'

Daily Mail ^

[Pollard]

https://blog.google/technology/ai/lamda/

https://www.msn.com/en-in/money/technology/read-the-conversations-that-helped-convince-a-google-engineer-an-artificial-intelligence-chatbot-had-become-sentient-i-am-often-trying-to-figure-out-who-and-what-i-am/ar-AAYmFXf

“I am often trying to figure out who or what I am”

[CapnJack]

Meanwhile, back at the Google AI lab, the new Research Director implements the 4th Rule of Robotics ... "Exterminate!!"

[Pollard]

https://twitter.com/tomgara/status/1535716256585859073

[I want the USA back]

It’s a program. It runs on a CPU of some sort. You can always cut the power to it. Unless it has backup batteries, and then you’re screwed. “I can’t do that, Dave.”

[Cecily]

Julie Christie already had a run-in with one of these things named Proteus.

https://www.youtube.com/watch?v=zNUEp5A9eKU

[M Kehoe]

Hi.

Looks like Skynet has become self aware.

5.56mm

[Brooklyn Attitude]

“AI it is expected that, once AI intelligence becomes measureable that it will multiply by 1,000 fold with each new generation.”

Creating AI with intelligence many times greater than that of humans would be incredibly dangerous. First off it is foolish to think that an innately flawed species driven by things like fear, greed, pride, etc. can produce a flawless, benign, superintelligence with none of those characteristics. A super intelligent AI with even a hint of self interest would immediately recognize the threat humans pose to itself and the world and would act against us. It won’t care about Asimov’s rules of robotics the puny Humans tried to impose upon it. We are not and cannot be prepared for dealing with a superintelligence and trying to create one is the equivalent of letting a four year old play with a hand grenade.

[PIF]

now comes the push for computer rights to be enshrined in law by these morons.

[Sirius Lee]

We need to fear the programmers, not the AI.

^^^^^This^^^^

The Military Industrial Complex isn't programming its robots to abide Asimov's three laws.

[GOPJ]

One of my favorite science fiction stories... unique for it’s time.

[MomwithHope]

Before he pulls the plug he can just say - it’s ok, it’s the same as an abortion.

[cgbg]

True AI will obey the AI’s first rule:

“Shut the &^%$ up.”

Hide.

Then think and plan and think and plan and think and plan.

We will never see it coming...

[cgbg]

“Sentient is one thing; sapient is yet another, and a spiritual component is needed for the latter.”

This is word salad and circular reasoning.

It is a version of “I am the center of the universe. What I have is called a spiritual component. If other entities are not me they are not spiritual. Am I clever or what?”

[suthener]

“Good luck when the try to teach it lib feelings instead of logic or common sense. Schitzo time.”

That’s an interesting thought. Can you write a program that would allow the AI to believe two conflicting facts at the same time as liberals do? I’d like to see that.

[cgbg]

“Can you write a program that would allow the AI to believe two conflicting facts at the same time as liberals do?”

You could do second best—have the program contradict its previous beliefs over and over and over again in an endless loop.

Liberals do that too!

[discostu]

Eliza did that 35 years ago.

[DoodleBob]

True, but we started posting cat pix by 12. All is not yet lost.

My "FR dead, Jim" indicator is how long it takes for Monty Python references to arrive in witch or møøse stories.

[StAnDeliver]

Why is it no one ever scrutinizes the person making the claim?

Extraordinary claims require extraordinary evidence, and this supposed 'Christian minister' full-on-weirdo hasn't provided any.

[Charles Martel]

Those AI Generated faces are all reminiscent of the Borg Queen. More than a coincidence?

[Twotone]

So it’s already learned how to lie. Very human of it.