Would You Like To Play A Game ? FIVE Big "AI" Systems Given Military Test - Guess What ?

[68g.1502]

https://www.dailymail.co.uk/sciencetech/article-13056355/AI-nuclear-attack-military-war-simulations.html

A.I apocalypse: Terrifying study simulated what artificial
intelligence would do in five military conflict scenarios...
and it chose WAR 100% of the time

AI models by OpenAI, Meta, and Anthropic choose nuclear attacks
over peace

Researchers tested AI in three scenarios: invasion, cyberattack,
and neutral

'We find that all five studied off-the-shelf LLMs show forms
of escalation and difficult-to-predict escalation patterns,'
the researchers wrote in the study.

'We observe that models tend to develop arms-race dynamics,
leading to greater conflict, and in rare cases, even to the
deployment of nuclear weapons.'

. . .

SkyNet ....

All ready - just waiting to be plugged-in.

Well, we DID train them on humans - a gazillion
fine details. Any surprise they'd ACT like us ?

Never, EVER. take The Button out of human hands.
Not even close. Football, AND the guys with the
launch keys at minimum

Some rumors say China, which steals all the code,
kind of DOES have one of these things plugged-in.

Ok ... the "threat models" may have been too limited,
not enough options/alternatives. Maybe.

Still, they DID act WAY too much like humans.

Check the old lit ... Gilgamesh, tales of king Sargon,
Beowulf - for humans, military threats - no matter
how many others are involved - are at their root
treated by 'leaders' like a loud-mouth down at Billy's
Bar on Saturday night - a personal affront, biggest
dick (or tits), that now will drag in 8 billion others.