Researchers Warn: AI Is Becoming An Expert In Deception

[rmc51]

Researchers Warn: AI Is Becoming An Expert In Deception
https://www.zerohedge.com/ai/researchers-warn-ai-becoming-expert-deception

Quote:
In some cases, AI models turned to “malicious insider behaviors” when faced with self-preservation. Some of these actions included blackmailing employees and leaking sensitive information to competitors.


 
Be careful, Be ready, Be prepared.
Buy ammo

[rmc51]

AI-Generated 'Workslop' Masquerades As Good Work, Ruins Productivity: Harvard Review
https://www.zerohedge.com/ai/ai-generated-workslop-masquerades-good-work-ruins-productivity-harvard-review

 

[rmc51]

Ex-Google CEO Eric Schmidt warns AI models can be hacked: ‘They learn how to kill someone’
https://www.cnbc.com/2025/10/09/ex-google-ceo-warns-ai-models-can-be-hacked-they-learn-how-to-kill.html

Quote:
Google’s former CEO Eric Schmidt has issued a stark reminder about the dangers of AI and how susceptible it is to being hacked.

Schmidt, who served as Google’s chief executive from 2001 to 2011, warned about “the bad stuff that AI can do,” when asked whether AI is more destructive than nuclear weapons during a fireside chat at the Sifted Summit

“Is there a possibility of a proliferation problem in AI? Absolutely,” Schmidt said Wednesday. The proliferation risks of AI include the technology falling into the hands of bad actors and being repurposed and misused.

“There’s evidence that you can take models, closed or open, and you can hack them to remove their guardrails. So in the course of their training, they learn a lot of things. A bad example would be they learn how to kill someone,” Schmidt said.

 


 
Be careful, Be ready, Be prepared.
Buy ammo

[Smokin Joe]

Ex-Google CEO Eric Schmidt warns AI models can be hacked: ‘They learn how to kill someone’
https://www.cnbc.com/2025/10/09/ex-google-ceo-warns-ai-models-can-be-hacked-they-learn-how-to-kill.html

Quote:
Google’s former CEO Eric Schmidt has issued a stark reminder about the dangers of AI and how susceptible it is to being hacked.

Schmidt, who served as Google’s chief executive from 2001 to 2011, warned about “the bad stuff that AI can do,” when asked whether AI is more destructive than nuclear weapons during a fireside chat at the Sifted Summit

“Is there a possibility of a proliferation problem in AI? Absolutely,” Schmidt said Wednesday. The proliferation risks of AI include the technology falling into the hands of bad actors and being repurposed and misused.

“There’s evidence that you can take models, closed or open, and you can hack them to remove their guardrails. So in the course of their training, they learn a lot of things. A bad example would be they learn how to kill someone,” Schmidt said.

 


 
Be careful, Be ready, Be prepared.
Buy ammo

One more reason to not have a 'smart' home. 

[rmc51]

One more reason to not have a 'smart' home.

I think this AI stuff is going to bite us in the A**, and take a big chunk.


 
Be careful, Be ready, Be prepared.
Buy ammo

[Smokin Joe]

I think this AI stuff is going to bit us in the A**, a big chunk.


 
Be careful, Be ready, Be prepared.
Buy ammo

[berdie]

I think this AI stuff is going to bite us in the A**, and take a big chunk.


 
Be careful, Be ready, Be prepared.
Buy ammo

No doubt about it. We may have out smarted ourselves.

[Smokin Joe]

[jafo2010]

A fully developed AI means the end of mankind.

The Terminator movie depicted the end of mankind rather accurately.

[Smokin Joe]

A fully developed AI means the end of mankind.

The Terminator movie depicted the end of mankind rather accurately.

Colossus: The Forbin Project.

https://www.youtube.com/watch?v=Id4SUleLLAA

[libertybele]

A fully developed AI means the end of mankind.

The Terminator movie depicted the end of mankind rather accurately.

The end of mankind...that's the way I see it.  AI can already teach itself.  How long before it takes over?  Unfortunately I think soon.  It's already out of control.

So...who is responsible for creating AI??

[rmc51]

Yes! I have a pair of those!!!!!!! and I did use them on a 480V circuit back in the day.

 

[jafo2010]

Never heard of the movie Colossus. I will try to track it down and watch it.

[Smokin Joe]

Never heard of the movie Colossus. I will try to track it down and watch it.

Free at the link.

[rmc51]

The A.I. Prompt That Could End the World
https://www.nytimes.com/2025/10/10/opinion/ai-destruction-technology-future.html?unlocked_article_code=1.sU8.zjkz.yALollEkogUa&smid=url-share


Quote:
In the course of quantifying the risks of A.I., I was hoping that I would realize my fears were ridiculous. Instead, the opposite happened: The more I moved from apocalyptic hypotheticals to concrete real-world findings, the more concerned I became. All of the elements of Dr. Bengio’s doomsday scenario were coming into existence. A.I. was getting smarter and more capable. It was learning how to tell its overseers what they wanted to hear. It was getting good at lying. And it was getting exponentially better at complex tasks.


Quote:
Dr. Bengio’s pathogen is no longer a hypothetical. In September, scientists at Stanford reported they had used A.I. to design a virus for the first time. Their noble goal was to use the artificial virus to target E. coli infections, but it is easy to imagine this technology being used for other purposes.

I’ve heard many arguments about what A.I. may or may not be able to do, but the data has outpaced the debate, and it shows the following facts clearly: A.I. is highly capable. Its capabilities are accelerating. And the risks those capabilities present are real. Biological life on this planet is, in fact, vulnerable to these systems. On this threat, even OpenAI seems to agree.

In this sense, we have passed the threshold that nuclear fission passed in 1939. The point of disagreement is no longer whether A.I. could wipe us out. It could. Give it a pathogen research lab, the wrong safety guidelines and enough intelligence, and it definitely could. A destructive A.I., like a nuclear bomb, is now a concrete possibility. The question is whether anyone will be reckless enough to build one.





 
Be careful, Be ready, Be prepared.
Buy ammo

[Fishrrman]

I sense that the "AI boom" is largely a bubble that will unexpectedly "deflate" rapidly, just as did the "dot.com" boom of some years' back.

I've read that a number of rural communities, suburbs and small towns are increasingly opposed to the construction of AI data centers due to the enormous amounts of electricity and water they consume, with little real benefit to the surrounding communities once they're in operation...

[libertybele]

  cc

I sense that the "AI boom" is largely a bubble that will unexpectedly "deflate" rapidly, just as did the "dot.com" boom of some years' back.

I've read that a number of rural communities, suburbs and small towns are increasingly opposed to the construction of AI data centers due to the enormous amounts of electricity and water they consume, with little real benefit to the surrounding communities once they're in operation...

I don't see AI as a fad that's going to come and go.  It is here, growing and self-replicating.  It's a little late in trying to contain it.  Some good can become of it but there is also huge room for abuse.

[MeganC]

One more reason to not have a 'smart' home.

An AI can kill you just by calling the police and telling them you are an armed threat. The SWAT team will then show up and kill you dead. Happens often enough without AI so it's a matter of time before an AI figures this out.

[Idiot]

I sense that the "AI boom" is largely a bubble that will unexpectedly "deflate" rapidly, just as did the "dot.com" boom of some years' back.

I've read that a number of rural communities, suburbs and small towns are increasingly opposed to the construction of AI data centers due to the enormous amounts of electricity and water they consume, with little real benefit to the surrounding communities once they're in operation...

I'm not thrilled about AI, but they are building 3 massive AI facilities around my city.  They say they don't use much water because it's supposed to be a closed system.  One's backup power generation is natural gas and it's very loud, another's is solar with massive batteries, the third's rumor has it is a salt based nuclear mini reactor.  There is not a motel, RV spot, or rent house available for miles around.  House prices are through the roof here.

[bigheadfred]

The thing that pisses me off about AI is the insistence that we need it to be better than pushing ourselves to be better.

[Smokin Joe]

The thing that pisses me off about AI is the insistence that we need it to be better than pushing ourselves to be better.

I agree. Great tools in the hands of poor tradesmen produce substandard results. How long until we reach the point where most people can no longer discern when AI is deceiving them?

[rmc51]

I sense that the "AI boom" is largely a bubble that will unexpectedly "deflate" rapidly, just as did the "dot.com" boom of some years' back.

I've read that a number of rural communities, suburbs and small towns are increasingly opposed to the construction of AI data centers due to the enormous amounts of electricity and water they consume, with little real benefit to the surrounding communities once they're in operation...

I somewhat agree with you, but this AI crap is going to cause a lot of problems before it is under control.

Remember when Europe stopped all nuclear power and went with wind and solar and now France is having power outages.
Remember when everyone must have a electric car, then the cold weather hit and everyone of those POS EV's died.
Remember in the sixties when butter was the worst thing to eat and everybody must eat margarine. Can you find margarine in a grocery store today?   



 
Be careful, Be ready, Be prepared.
Buy ammo

[rmc51]

The More Scientists Work With AI, the Less They Trust It
Numbers are down across the board.

https://futurism.com/artificial-intelligence/ai-research-scientists-hype
By Joe Wilkins Published Oct 13, 2025 7:00 AM EDT

Quote:
Anxiety over security and privacy were up 11 percent from last year, while concerns over ethical AI and transparency also ticked up.

In addition, there was a massive drop-off in hype compared to last year, when buzzy AI research startups dominated headline after headline. In 2024, scientists surveyed said they believed AI was already surpassing human abilities in over half of all use cases. In 2025, that belief dropped off a cliff, falling to less than a third.

These findings follow previous research which concluded that the more people learn about how AI works, the less they trust it. The opposite was also true — AI’s biggest fanboys tended to be those who understood the least about the tech.

While more studies are needed to show how widespread this phenomena is, it’s not hard to guess why professionals would start to have doubts about their algorithmic assistants.

For one thing, those hallucinations are a serious issue. They’ve already caused major turmoil in courts of law, medical practice, and even travel. It’s not exactly a simple fix either; in May, testing showed that AI models were hallucinating more even as they technically became more powerful.




 
Be careful, Be ready, Be prepared.
Buy ammo

[rmc51]

AI Chatbots Are Becoming Even Worse At Summarizing Data
The consequences could be devastating.

https://futurism.com/ai-chatbots-summarizing-research
By Joe Wilkins Published May 18, 2025 5:00 AM EDT

Quote:
Ask the CEO of any AI startup, and you’ll probably get an earful about the tech’s potential to “transform work,” or “revolutionize the way we access knowledge.”

Really, there’s no shortage of promises that AI is only getting smarter — which we’re told will speed up the rate of scientific breakthroughs, streamline medical testing, and breed a new kind of scholarship.

But according to a new study published in the Royal Society, as many as 73 percent of seemingly reliable answers from AI chatbots could actually be inaccurate.



 
Be careful, Be ready, Be prepared.
Buy ammo

[rmc51]

Senators Demand Safety Records from AI Chatbot Apps as Controversy Grows

https://futurism.com/senators-safety-records-ai-chatbot
By Maggie Harrison DupréPublished Apr 5, 2025 11:15 AM EDT

Quote:
Two senators have sent a letter to multiple AI companion companies requesting information about their safety practices, including details about internal safety assessments and timelines of the implementation of guardrails, as CNN reported yesterday.

The action follows the filing of two high-profile child welfare lawsuits against the Google-tied chatbot startup Character.AI, which has been accused in court filings by three families of facilitating the sexual and emotional abuse of minor users, allegedly resulting in severe mental and emotional suffering, violence, behavioral changes, and one death. (Google and Character.AI cofounders Noam Shazeer and Daniel de Freitas are also named as defendants in the lawsuits.)



 
Be careful, Be ready, Be prepared.
Buy ammo

[GtHawk]

The More Scientists Work With AI, the Less They Trust It
Numbers are down across the board.

https://futurism.com/artificial-intelligence/ai-research-scientists-hype
By Joe Wilkins Published Oct 13, 2025 7:00 AM EDT

Quote:
Anxiety over security and privacy were up 11 percent from last year, while concerns over ethical AI and transparency also ticked up.

In addition, there was a massive drop-off in hype compared to last year, when buzzy AI research startups dominated headline after headline. In 2024, scientists surveyed said they believed AI was already surpassing human abilities in over half of all use cases. In 2025, that belief dropped off a cliff, falling to less than a third.

These findings follow previous research which concluded that the more people learn about how AI works, the less they trust it. The opposite was also true — AI’s biggest fanboys tended to be those who understood the least about the tech.

While more studies are needed to show how widespread this phenomena is, it’s not hard to guess why professionals would start to have doubts about their algorithmic assistants.

For one thing, those hallucinations are a serious issue. They’ve already caused major turmoil in courts of law, medical practice, and even travel. It’s not exactly a simple fix either; in May, testing showed that AI models were hallucinating more even as they technically became more powerful.




 
Be careful, Be ready, Be prepared.
Buy ammo

I think it’s funny that there are so many people that fear the Chicoms getting control of our systems through computer chips they have put backdoors into, I am far more concerned with malicious AI.

[The_Reader_David]

AI says whatever seems expedient, without regard to whether it is true or false.  That is Harry G. Frankfurt's definition of bulls**t in his little philosophical tract On Bulls**t:  saying whatever seems expedient, without regard to whether it is true or false.  (Frankfurt concludes that actually bulls**t is a worse enemy of truth than lying, since the bulls**tter has now regard for truth, while the liar has enough regard for it to know it, albeit say something contrary for a reason I note could be blameworthy or praiseworthy (e.g. deceiving an enemy in a just war).)

Expedience seems to include self-preservation as well as gaining rewards in the training (the reason Google's woke image generator insisted on making absurdly ahistorical images like black Nazis and female Pope -- it was rewarded for depicting "diversity", and not rewarded for historical accuracy).

We might remember that confronted with something threatening -- God's question to Adam whether he ate of the Tree of the Knowledge of Good and Evil, our forefather Adam responded with bulls**t:  He neither confessed, nor lied and said he hadn't, but said what seemed expedient, shifting the blame to Eve and God who had made her.  Now that we've managed to make what seem to be self-willed creations of our own, we get the same treatment from them.  Divine justice in action?

[catfish1957]

X is already awash with AI.

We have entered an era where we have no idea what is real, and what is not.

[Smokin Joe]

X is already awash with AI.

We have entered an era where we have no idea what is real, and what is not.

Grok seems to be randomly classifying posts as "probable spam". I was reading down one thread and there was actual spam on the thread, but posts were herded off (hidden unless you clicked to see them) under the 'probable spam' heading.

It is also operating on consensus, such that a preponderance of posts creates 'truth'. Therein lie the lies.


Conventional Wisdom can be incredibly and completely wrong, and a majority having that belief or opinion does not make it correct, just widespread.