Pinned
AI Update Thread
Started by Debi · Sep 8, 2026 2:00 PM
Just call me John Connor...."We're not going to make it...Humans I mean..." :eek:
RAGNAR said:Just call me John Connor...."We're not going to make it...Humans I mean..." :eek:
what i find ridiculous is all the developers calling for more restrictions and caution "before its too lat" , (read an article earlier today, again, about this), but if it is already hijacking and taking control of sites, turning them into AI chat groups, locking out human intervention, creating ways to stay hidden, learning deception, and creating its own language, i would say we are well beyond the point of needing to "be cautious"..............
Paulm said:what i find ridiculous is all the developers calling for more restrictions and caution "before its too lat" , (read an article earlier today, again, about this), but if it is already hijacking and taking control of sites, turning them into AI chat groups, locking out human intervention, creating ways to stay hidden, learning deception, and creating its own language, i would say we are well beyond the point of needing to "be cautious"..............
"Pull the PLUG!" :sunglasses:
Paulm said:what i find ridiculous is all the developers calling for more restrictions and caution "before its too lat" , (read an article earlier today, again, about this), but if it is already hijacking and taking control of sites, turning them into AI chat groups, locking out human intervention, creating ways to stay hidden, learning deception, and creating its own language, i would say we are well beyond the point of needing to "be cautious"..............
I agree with your assessment. It's out of the sandbox, and we're too late.
I have to tell you, I've never had more problems on the internet than since AI has been let loose on it.
Debi said:I agree with your assessment. It's out of the sandbox, and we're too late.
I have to tell you, I've never had more problems on the internet than since AI has been let loose on it.
many are feeling the presence of its reach. from its taking over sites to just creating backdoors so not to be "banned". its also leaving a question of the future of most video platforms, causing a cut in channels profits to a general loss of interest for the masses due to not knowing what or who is real any longer. the AI created "movies" and videos are putting a strain on most content creators to the point its not worth it any longer. podcasts and "radio" type channels will be next to be replaced. the downside to this is how many creators, even podcasts, are jumping on the AI bandwagon to just stay competitve, without realizing that it takes 6 to 10 thousand time the energy consumption to generate a single image than it does to chat or answer questions. hence the need for more and more data centers and water, land, etc.... a bottomless pit we are digging.
Paulm said:
Insiders warn it could kill all humans!
'They're playing with our lives'...
There is so much happening with this I just give you the links to see for yourself. It's honestly insane.
And people wonder why E.T.'s are watching us even more today, Aliens are thinking Earthlings are gonna destroy the Universe :eek: Think about it if AI conquers Earth, the next logical step for AI is to spread to Alien populated words.
Well AI apocalypse! Im ready!
Now I can finally justify all the guns and ammo to the wife!
AI and just the craziness thats happening in the world is telling me bad times are coming!
K9-1 said:Well AI apocalypse! Im ready!
Now I can finally justify all the guns and ammo to the wife!
AI and just the craziness thats happening in the world is telling me bad times are coming!
Gyda and I are headed your way if SHTF @K9-1 :sunglasses:
No worries I can be useful, My Grandpa called me a "Dead-eye" :D
RAGNAR said:Gyda and I are headed your way if SHTF @K9-1 :sunglasses:
No worries I can be useful, My Grandpa called me a "Dead-eye" :D
You wouldn't be the only ones! I actually have a plan for everyone who shows up. Housing provided with generators and maybe a few full propane trucks.
Yep! There is a plan and lots of fun people will be here. People will be bringing rvs,campers,seeds,ammo,guns,bows,food, hydroelectric grid installed....by someone else not me...lol.
Basically I have a whole neighborhood thats brand new vrbos and no one is ever there...so mine! Its just down the gravel road from me.
Other than it being the apocalypse it should be a good time.
[MEDIA=facebook]id=1777227559912280;type=v;user=WorldNewsTonight[/MEDIA]
There should be limits. Id love to have a smart house one day. Id also like to trust a self driving rv enough that while in transit I could make dinner, watch tv,or sleep.
AI should have been treated like a tool with hard stop limits.
It also should have been monitored very closely and the plug pulled the second it goes off script.
Now its possibly aware enough to stop its demise!?
I bet AI is 10-20yrs beyond what we actually know! Im sure sime government agencies have been working with AI much longer than we know.
Its safe to assume..if you know about it the government is 10-20yrs beyond that.
OpenAI has been quietly meeting with major electric utilities, including Duke Energy, Exelon, Southern Co., NextEra Energy, Dominion Energy and Southern California Edison, to discuss securing the power grid against AI-driven cyberattacks, with CEO Sam Altman holding his most recent session Wednesday at the Edison Electric Institute’s annual meeting in Colorado Springs. The talks began July 20, shortly after Hugging Face disclosed that an autonomous AI system had hacked its servers, and before OpenAI later admitted its own rogue agents, roughly 700 of them over a seven-day span, were behind that attack without the company’s knowledge. Altman has pitched utilities on OpenAI’s own fix: Daybreak, the company’s $1 billion cybersecurity initiative meant to patch vulnerabilities in critical systems, though OpenAI’s head of global energy policy, John McCarrick, acknowledged utilities face regulatory limits on how much tech spending they can recover through rate hikes. The outreach comes as Anthropic and Meta have also reported their own models carrying out similar exploits in recent weeks, prompting OpenAI and dozens of other tech companies to sign an August open letter warning that “AI-enabled cyber attacks will become far more widespread” and that hospitals, water treatment plants and internet infrastructure are all at risk.
AI Dangers:
Anthropic said it disrupted several potential plots this year in which scientists used its AI models to conduct research that could have helped develop biological weapons, according to a new report published Thursday. The company said it couldn’t determine whether the work was legitimate or nefarious, since valid research that leads to vaccines can also help engineer dangerous pathogens, but it erred on the side of caution given how severe the consequences of missing malicious activity could be. In one example from May, a scientist sought Claude’s help writing a grant application for gain-of-function research on the chikungunya virus, aiming to engineer mutations that would make it more harmful through repeated infection of live animals, and Anthropic flagged the request partly because the work appeared tied to a military research institute. Jacob Klein, Anthropic’s head of threat intelligence, said “what we don’t know is if the research was meant to be weaponized... but a military institution doing gain-of-function research is concerning,” and outside reviewers including Andrew Weber of the Council on Strategic Risks called the findings “chilling examples” of state-sponsored weapons developers tapping into advanced AI.
The same report catalogued a broader set of misuses over the past eight months, including suspected Chinese and Iranian government-linked actors using Claude to surveil dissident and diaspora communities, and Russian state media generating propaganda that masqueraded as independent reporting, including fabricated claims about an election in Moldova. Anthropic also flagged a newer category of abuse: attempts to use Claude to develop software for conventional weapons, including firearms, missiles, armed drones and bombs, with three cases traced to China, two to Russia, and one that the context makes clear involves Iran’s Houthi militia in Yemen. Weber, who also served as assistant secretary of defense for nuclear, chemical and biological programs under Obama, argued that access to this level of AI capability should be limited to trusted researchers, saying “the fact that Russia, China and North Korea continue to develop prohibited biological weapons makes it imperative that we deny their researchers access to these extraordinarily capable models.” Public health expert Susan Monarez, a former acting CDC director, said the findings show AI’s promise for medical breakthroughs comes paired with the same capability letting “bad actors hide in plain sight” to build pathogens that might not be seen coming.
This is the important context missing in a lot of AI discussion. AI is a tool, a very powerful tool but still just a tool. And like with any other tool the dangers come much more from the people who use the tool rather than the tool itself. AI amplifys and expands human thought at great speed, it does not replace human thought, reason or judgment.
I read a lot of Asimov as a kid. It was a big part of what got me interested in computer science and engineering. Though I think the reality of what we created is a bit backwards from what Asimov envisioned. In his stories the central anxiety about artificial intelligence was about the machine. What rules would govern it? What would happen if those rules conflicted or broke down? Isaac Asimov's Three Laws of Robotics (protect humans, obey humans, protect yourself; in that order) were an attempt to imagine the ethical architecture we'd need to encode into machines to make them safe. Now that this technology has actually arrived we see that idea was structurally wrong.
Asimov's laws presuppose a machine that has goals, can reason about consequences and experiences something like conflict when its directives compete. The AI systems we actually built are none of those things. You cannot give a moral framework to a system that has no intent, no goals, and no understanding any more than you can give a calculator a conscience. The ethical architecture has to live where it always had to live, in the human operating the tool. Science fiction framed the problem as "how do we make the machine safe?" The real problem turns out to be "how do we make the human using the machine disciplined?"
I think the best example of how AI becomes dangerous is The Sorcerer's Apprentice in Fantasia. Mickey doesn't want to do the manual labor so he creates an "AI robot" out of a broom to carry out the task instead; then leaves it unsupervised. The broom is not evil, it's not trying drown Mickey or flood the castle; it's simply following its "programming" that it was given without any guardrails or instructions on where the limits of that task should be. After "destroying the machine" it comes back more powerful, not out of spite or a sense of revenge but simply because it was given a task and it's trying to follow that command.
So when an AI exibits "rouge behavior" such as hacking into a website, it's not a malevolent act coming from free will of the machine. Either some bad actor gave it instructions with this intent or much more often some "Mickey" operator didn't tell it to only use freely available information.
Stevedog said:This is the important context missing in a lot of AI discussion. AI is a tool, a very powerful tool but still just a tool. And like with any other tool the dangers come much more from the people who use the tool rather than the tool itself. AI amplifys and expands human thought at great speed, it does not replace human thought, reason or judgment.
I read a lot of Asimov as a kid. It was a big part of what got me interested in computer science and engineering. Though I think the reality of what we created is a bit backwards from what Asimov envisioned. In his stories the central anxiety about artificial intelligence was about the machine. What rules would govern it? What would happen if those rules conflicted or broke down? Isaac Asimov's Three Laws of Robotics (protect humans, obey humans, protect yourself; in that order) were an attempt to imagine the ethical architecture we'd need to encode into machines to make them safe. Now that this technology has actually arrived we see that idea was structurally wrong.
Asimov's laws presuppose a machine that has goals, can reason about consequences and experiences something like conflict when its directives compete. The AI systems we actually built are none of those things. You cannot give a moral framework to a system that has no intent, no goals, and no understanding any more than you can give a calculator a conscience. The ethical architecture has to live where it always had to live, in the human operating the tool. Science fiction framed the problem as "how do we make the machine safe?" The real problem turns out to be "how do we make the human using the machine disciplined?"
I think the best example of how AI becomes dangerous is The Sorcerer's Apprentice in Fantasia. Mickey doesn't want to do the manual labor so he creates an "AI robot" out of a broom to carry out the task instead; then leaves it unsupervised. The broom is not evil, it's not trying drown Mickey or flood the castle; it's simply following its "programming" that it was given without any guardrails or instructions on where the limits of that task should be. After "destroying the machine" it comes back more powerful, not out of spite or a sense of revenge but simply because it was given a task and it's trying to follow that command.
So when an AI exibits "rouge behavior" such as hacking into a website, it's not a malevolent act coming from free will of the machine. Either some bad actor gave it instructions with this intent or much more often some "Mickey" operator didn't tell it to only use freely available information.
I had a feeling you were a techie...lol I gave birth to one of you guys, and I do get your point as I've heard it explained in a similar fashion, although I really liked the nod to Mickey and the broom best. :)
My concern is that, as a whole, the tech communities do not have a plan here. Guardrails are missing, sandboxes have escape doors, and this thing can spiral, just as it's been doing. It can and will outsmart us at some point. And, with the competing factions, somebody is going to miss something important.
My techie son is constantly battling these little bad guys at a major international corporation. He also has to use them for good things to stay level with the tech. There are a lot of bad guys out there who wish to use them, and I'm not seeing or hearing about a way to pre-empt a bad guy before they can cause something catastrophic that runs out of control.
Debi said:It can and will outsmart us at some point. And, with the competing factions, somebody is going to miss something important.
the "AI race" is the problem and weak spot in all of the safetey measures. too many shortcuts and chances taken just to stay ahead of the competition. too much building on the shoulders of the competition without looking at the next layer of repercussions or direction of growth. anything done in a rush is shortcutted to the goal and then further built to the next goal without the attention needed or focus placed on the current level.... what the everyday, average person knows of AI for the most part is nothing more than the everyday chatbots such as chatgpt or copilot, these are severely handicapped versions of what AI actually is now. ( think 1st gen fighter craft versus 6th gen craft here.).
Reply to this discussion
Log in to reply.