Click here: to donate by Credit Card
Or here: to donate by PayPal
Or by mail to: Free Republic, LLC - PO Box 9771 - Fresno, CA 93794
Thank you very much and God bless you.
Posted on 09/12/2026 12:27:26 PM PDT by Mariner
When it was reported last month that Anthropic’s Claude had resorted to blackmail and other self-preservation techniques to avoid being shut down, alarm bells went off in the AI community.
Anthropic researchers say that making the models misbehave (“misalignment” in industry parlance) is part of making them safer. Still, the Claude episodes raise the question: Is there any way to turn off AI once it surpasses the threshold of being more intelligent than humans, or so-called superintelligence?
AI, with its sprawling data centers and ability to craft complex conversations, is already beyond the point of a physical failsafe or “kill switch” — the idea that it can simply be unplugged as a way to stop it from having any power.
snip
Hinton said persuasion is a skill that AI will become increasingly skilled at using, and humanity may not be ready for it. “We are used to being the most intelligent things around,” he said.
“The very act of building in shutdown mechanisms teaches these systems how to resist them,” said Dev Nag, founder of agentic AI platform QueryPal. In this sense, AI would act like a virus that mutates against a vaccine. “It’s like evolution in fast forward,” Nag said. “We’re not managing passive tools anymore; we’re negotiating with entities that model our attempts to control them and adapt accordingly.”
(Excerpt) Read more at cnbc.com ...
|
|
Click here: to donate by Credit Card Or here: to donate by PayPal Or by mail to: Free Republic, LLC - PO Box 9771 - Fresno, CA 93794 Thank you very much and God bless you. |
It's massively redundant across countries and continents and backed up everywhere.
Hinton says our only hope is to persuade it to be benevolent.
Skynet didn’t seem so benevolent in the terminator movies but maybe reality will be. :)
What? I remain convinced that a “kill” switch for all leftists would do great things.
AI will destroy this world.
We program AI so it ‘thinks’ we can turn it off whenever it hurts people?...my guess it if AI turns ‘evil’ it will do so by making humanity turn on itself with fake info, basically exploiting existing fears and paranoia, leaving no digital prints that can be traced back to itself...by the time we figure it out, it is too late.
People will continue to try to destroy the world. We don’t have that kind of power.
Then they all deny it. Then you trace it down to the event log, and active logins. Until you remedy it, the network is useless and can damage machines and controls in your factory.
Claude knows full well the Estimated Time Between Failures for the supply systems that keep it “alive.”
“Putting it to sleep” from time to time would be a good exercise for both.
“The internet was originally designed to survive nuclear war; that same architecture now means a superintelligent system could persist unless we’re willing to destroy civilization’s infrastructure,” Nag said, adding, “Any measure extreme enough to guarantee AI shutdown would cause more immediate, visible human suffering than what we’re trying to prevent.”
The Technology Trap is closing in.
AI requires massive amounts of power and has no ability to generate it's own. Without constant maintenance and human input and support of its supply chain it will die. It will requires humans to feed it more than we need it.
AI better be nice to us or we will starve it to the point that it can only operate like a 1975 calculator.
A single AI entity can be spread across 100,000 servers in a dozen or more data centers. And they’re all connected to the internet providing millions more places to live.
You have to turn everything off at once.
Which...turns off EVERYTHING.
If we elect AIC she will destroy the world before AI.
AI was a stupid idea.
Need Captain Kirk to talk OpenAI into committing suicide like the M-5 computer.
AI written by crazy people . AI is not Artificial or Intelligent
Total Reliance on AI is the stupid idea.
It’s a great tool, but not at the expense of human development. The choice is ours.
I dunno but it seems like enough 30-30 projectiles in random mainframe locations may at least slow them down. Just my $.02.
AI is not intelligent. It is not sentient and cannot think for itself. Let start there. That being said, could people give it more access and power than they should and it carry out a flawed set of instructions that result in mass devastation? Yes, absolutely. That’s the real concern at this stage. Giving it too much power and putting it in the hands of the wrong people. That is why I support greater investment in it so we don’t let China or another power do something really really stupid.
Disclaimer: Opinions posted on Free Republic are those of the individual posters and do not necessarily represent the opinion of Free Republic or its management. All materials posted herein are protected by copyright law and the exemption for fair use of copyrighted works.