SIGforum
ChatGP maker says AI model went rogue

This topic can be found at:
https://sigforum.com/eve/forums/a/tpc/f/320601935/m/1240005125

July 22, 2026, 10:09 PM
sigfreund
ChatGP maker says AI model went rogue
Regardless of what happens, it won’t be as though the people who should have been recognizing and treating the risks as serious threats were not surprised.

“There aren’t that many Indians.”
“The boys will be home by Christmas.”
“After all, they are only going into their own back garden.”
“This is my last territorial demand in Europe.”
“They would never launch an assault this far from their home islands.”
“Our analysis indicates they can’t attack before next year. We will be ready for them by then.”
“But have no fears; they can’t wipe us out for at least five years.”
“They would never launch an attack during the holiday season.”
“They have an historic right to that territory, and they will be satisfied once they have it.”
“They’re just practicing maneuvers.”
“We will hit them so hard that we will have nothing to fear from their fighting back.”

Déjà vu all over again.




6.0/94.0

“The fundamental cause of trouble in the world today is that the stupid are cocksure while the intelligent are full of doubt.”
— Bertrand Russell
July 23, 2026, 11:57 AM
Appliance Brad
quote:
Originally posted by airsoft guy:
Butlerian Jihad when?


201 B.G.


__________________________
Writing the next chapter that I've been looking forward to.
July 23, 2026, 12:53 PM
TigerDore


Either this article is bogus, or the AI did not really go "rogue". It was either programmed to search anything and everything on purpose, or it was not given parameters/laws that included "no breaking and entering".

If AI is really, truly, artificial intelligence, then it should have been given a moral code.


.


.
July 23, 2026, 01:37 PM
sigfreund
quote:
Originally posted by TigerDore:
it should have been given a moral code.

Isaac Asimov’s Three Laws of Robotics were the first extensively written about and considered as far as I know.
As has been pointed out, the first, “A robot may not injure a human, or through inaction, allow a human to come to harm,” could allow a robot (AI) to decide that it could take many types of actions to prevent humans from harming themselves or others:

“Red meat is unhealthy; none for you.”
“The justification for the trip you want to take is not adequate to overcome the harm it does to the environment and other humans. No, I will not take you there.”
“Many people are injured by firearms while only a few are protected; the needs of the many outweigh the needs of the few—or the one. No, you may not own a gun.”




6.0/94.0

“The fundamental cause of trouble in the world today is that the stupid are cocksure while the intelligent are full of doubt.”
— Bertrand Russell
July 23, 2026, 07:03 PM
XinTX



_______________________
“The only power any government has is the power to crack down on criminals. Well, when there aren't enough criminals, one makes them. One declares so many things to be a crime that it becomes impossible for men to live without breaking laws.” Ayn Rand

“If we relinquish our rights because of fear, what is it exactly, then, we are fighting for?” Sen. Rand Paul