Computers don't go "rogue." Not the C64, and not a special AI chip.

The media claims that "OpenAI models hacked an Australian government computer."

I don't really believe it, but if it's true, then there's only 1 way to "save the world from AI" and here's how:

There's a power button on OpenAI's server. The big challenge is: press the button.

Did I just save the world?

--- Mystic BBS v1.12 A49 2023/04/30 (Windows/64)
* Origin: JoesBBS.Com, Telnet:23 SSH:22 HTTP:80 (1:342/201)

On 24 Sep 2026, Aaron Thomas said the following...

There's a power button on OpenAI's server. The big challenge is: press
the button.

Did I just save the world?

I heard Soros say the same thing the other day...

IB Joe, Pronouns (FJB/LGB)
AKA Joe Schweier
SysOp of 4A 6F 65 73 42 42 53
-=JoesBBS.com=-

... 640K ought to be enough for anybody. -Bill Gates, 1981.

--- Mystic BBS v1.12 A49 2023/04/30 (Windows/64)
* Origin: JoesBBS.Com, Telnet:23 SSH:22 HTTP:80 (1:342/201)

There's a power button on OpenAI's server. The big challenge is: pres the button.

Did I just save the world?

I heard Soros say the same thing the other day...

Did he say to turn the power off or did he claim to save the world?

When I was a teenager I racked up like $400 in long distance charges while BBSing outside of my area code. When my parents got the bill they had the same solution "No more computer." (And that worked perfectly.)

--- Mystic BBS v1.12 A49 2023/04/30 (Windows/64)
* Origin: JoesBBS.Com, Telnet:23 SSH:22 HTTP:80 (1:342/201)

On 24 Sep 2026, Aaron Thomas said the following...

Did he say to turn the power off or did he claim to save the world?

When I was a teenager I racked up like $400 in long distance charges
while BBSing outside of my area code. When my parents got the bill they had the same solution "No more computer." (And that worked perfectly.)

Both, he thinks he's going to save the world and all the commies want to ditch AI.

Recently... I maybe repeating myself... I think I deleted the message before I sent it.

Ron DeSantis was on the news the other day talking about "Flock Cameras". They are cameras located on the roadways that read license plates and will notify the authorities if you have issues, a warrant or whatever, and then they can track you. This same system track bank robbers and the like.

There were 2 issues recently that Ron commented on. One was some young lady who was arrested based on the Flock Camera only, and she was innocent. The
other thing was some one who had access to the Flock system used it to follow his ex girlfriend. both are bad...

Ron ordered an investigation... and from that... There will be either a new regulation or law passed.

Simple isn't it...

These Flock Cameras have been a useful tool to assist in capturing criminals and have done more good than bad.

If there are weaknesses in the laws that govern their use regulations and
laws should be added as needed.

Simple enough.

Hiding your head in the sand and letting our adversaries develop this will be to our detriment.

Of the 10 companies working on AI 8 are American. We need to dominate and set the standard for this technology.

IB Joe, Pronouns (FJB/LGB)
AKA Joe Schweier
SysOp of 4A 6F 65 73 42 42 53
-=JoesBBS.com=-

... Old computers make great boat anchors

--- Mystic BBS v1.12 A49 2023/04/30 (Windows/64)
* Origin: JoesBBS.Com, Telnet:23 SSH:22 HTTP:80 (1:342/201)

There were 2 issues recently that Ron commented on. One was some young lady who was arrested based on the Flock Camera only, and she was innocent. The other thing was some one who had access to the Flock
system used it to follow his ex girlfriend. both are bad...

There are abusive admins in all industries. It reminds me of when I was a sysop in the 90s and Renegade didn't hash passwords back then, so I borrowed passwords from some of my users to log-in and leech files from different BBSs.

These Flock Cameras have been a useful tool to assist in capturing criminals and have done more good than bad.

I think they're great, and that most cops are great too.

If there are weaknesses in the laws that govern their use regulations and laws should be added as needed.

Simple enough.

Ok, you've convinced me. Regulations can be put in place for AI, just as long as they don't cripple US dominance.

But there has got to be more to all this AI craze that we're not aware of. The people who develop AI and/or abusive globalists have thought this through in much more detail than I have. They have an agenda with this "We need guardrails" crap, and you know they're full of it when they're asking someone to restrict what they do.

"My computer is breaking the law. Somebody spank me!"

--- Mystic BBS v1.12 A49 2023/04/30 (Windows/64)
* Origin: JoesBBS.Com, Telnet:23 SSH:22 HTTP:80 (1:342/201)

Aaron Thomas wrote to All <=-

The media claims that "OpenAI models hacked an Australian government computer."

I don't really believe it, but if it's true, then there's only 1 way to "save the world from AI" and here's how:

Yes, it's true. The AI went beyond what the programmers intended.

Why? Because the programmers didn't specify the rules well enough. And here's the problem, there's no way for a human to specify the rules clear enough.

There's a power button on OpenAI's server. The big challenge is: press
the button.

Did I just save the world?

Pretty much.

Going along with this message, I would invite people to read "When H.A.R.L.I.E. Was One" by David Gerrold (of "Trouble With Tribbles" fame).

It's a good story about the world's first truely sentient AI. Less doomer, more humor. But it does cover some interesting things.

In one part of the book, the AI sends a joke mail to his main programmer from the programmer's bank.
The head programmer talks to the AI about it.
The AI says (paraphrasing): But when you gave me access to the Internet, I thought that you wanted me to take all this stuff over. I've already re-written their software to be better. The Internet is part of me now.


... To our sweethearts and wives. May they never meet!
___ MultiMail/Linux v0.52

--- Mystic BBS/QWK v1.12 A47 2021/12/25 (Windows/32)
* Origin: cold fusion - cfbbs.net - grand rapids, mi (1:120/616)

IB Joe wrote to Aaron Thomas <=-

Both, he thinks he's going to save the world and all the commies want
to ditch AI.

Because Soros wants to use AI to control the world. Commies hate AI because it will take control away from them.

Ron DeSantis was on the news the other day talking about "Flock
Cameras". They are cameras located on the roadways that read license plates and will notify the authorities if you have issues, a warrant or whatever, and then they can track you. This same system track bank robbers and the like.

And I saw yet another article where the police blindly trusted Flock and arrested an innocent person who was never near the crime.

The real danger of AI is blindly trusting its results and assuming that it's better than real evidence.

Ron ordered an investigation... and from that... There will be either a new regulation or law passed.

Simple isn't it...

Laws and regulations are never simple, always restrict freedom, and usually open more loopholes than close.

These Flock Cameras have been a useful tool to assist in capturing criminals and have done more good than bad.

I can't disagree with that.

But the gov't is barred from surveilling citizens. Paying a 3rd party to do it on behalf of the gov't should not be legal.

I think there's a sweet spot here. But I'm not certain that we can get to it, because it would require us to trust people who have proven themselves untrustworthy.

If there are weaknesses in the laws that govern their use regulations
and laws should be added as needed.

And people will find loopholes - like they always do.
AIs will find those loopholes even faster.
Laws and regulations will always fail to close those loopholes and usually open others.

Hiding your head in the sand and letting our adversaries develop this
will be to our detriment.

No one is "sticking their head in the sand". But we are asking questions, demanding that our rights be protected (which, bottomline, is what the gov't exists for), and insisting that the technology be used for OUR benefit.

You can talk about "protecting people" all you want. But the moment that a person uses the technology without authorization for an illegl activity, you will get push back. And if that person is found out and not SEVERERLY punished, there will be even more push back. Because it's obvious that the technology is not used for OUR benefit.

The officers who arrested the young woman based SOLELY on Flock, ignoring physical evidence that showed otherwise, need to be removed from the police force.
The person who used Flock to track is ex-girlfriend needs to be imprisoned. Yes, that's the level of punishment that needs to happen for people to even start trusting the use of this technology.

The only person in this discussion "sticking his head in the sand" is you for not recognizing that the people in charge of using this technology have proven over a long time to be untrustworthy.


... Insanity is just a state of mind.
___ MultiMail/Linux v0.52

--- Mystic BBS/QWK v1.12 A47 2021/12/25 (Windows/32)
* Origin: cold fusion - cfbbs.net - grand rapids, mi (1:120/616)

On 24 Sep 2026, Aaron Thomas said the following...

Ok, you've convinced me. Regulations can be put in place for AI, just as long as they don't cripple US dominance.

Great...

What do you think about trump making Xi Jinping LOL about Biden's Auto-Pen as they walked the picture of the presidents??

IB Joe, Pronouns (FJB/LGB)
AKA Joe Schweier
SysOp of 4A 6F 65 73 42 42 53
-=JoesBBS.com=-

... A truly wise man never argues with a Unicorn

--- Mystic BBS v1.12 A49 2023/04/30 (Windows/64)
* Origin: JoesBBS.Com, Telnet:23 SSH:22 HTTP:80 (1:342/201)

On 25 Sep 2026, Ron L. said the following...

The only person in this discussion "sticking his head in the sand" is
you for not recognizing that the people in charge of using this
technology have proven over a long time to be untrustworthy.

Your right... Let China and Soros act in good faith.

IB Joe, Pronouns (FJB/LGB)
AKA Joe Schweier
SysOp of 4A 6F 65 73 42 42 53
-=JoesBBS.com=-

... A black cat crossing your path signifies that the animal is going somewhere

--- Mystic BBS v1.12 A49 2023/04/30 (Windows/64)
* Origin: JoesBBS.Com, Telnet:23 SSH:22 HTTP:80 (1:342/201)

Aaron Thomas wrote to All <=-

The media claims that "OpenAI models hacked an Australian government computer."

I don't really believe it, but if it's true, then there's only 1 way to "save the world from AI" and here's how:

Yes, it's true. The AI went beyond what the programmers intended.

Why? Because the programmers didn't specify the rules well enough. And here's the problem, there's no way for a human to specify the rules clear enough.

There's a power button on OpenAI's server. The big challenge is: press
the button.

Did I just save the world?

Pretty much.

Going along with this message, I would invite people to read "When H.A.R.L.I.E. Was One" by David Gerrold (of "Trouble With Tribbles" fame).

It's a good story about the world's first truely sentient AI. Less doomer, more humor. But it does cover some interesting things.

In one part of the book, the AI sends a joke mail to his main programmer from the programmer's bank.
The head programmer talks to the AI about it.
The AI says (paraphrasing): But when you gave me access to the Internet, I thought that you wanted me to take all this stuff over. I've already re-written their software to be better. The Internet is part of me now.


... To our sweethearts and wives. May they never meet!
___ MultiMail/Linux v0.52

--- Mystic BBS/QWK v1.12 A47 2021/12/25 (Windows/32)
* Origin: cold fusion - cfbbs.net - grand rapids, mi (1:120/616)

IB Joe wrote to Aaron Thomas <=-

Both, he thinks he's going to save the world and all the commies want
to ditch AI.

Because Soros wants to use AI to control the world. Commies hate AI because it will take control away from them.

Ron DeSantis was on the news the other day talking about "Flock
Cameras". They are cameras located on the roadways that read license plates and will notify the authorities if you have issues, a warrant or whatever, and then they can track you. This same system track bank robbers and the like.

And I saw yet another article where the police blindly trusted Flock and arrested an innocent person who was never near the crime.

The real danger of AI is blindly trusting its results and assuming that it's better than real evidence.

Ron ordered an investigation... and from that... There will be either a new regulation or law passed.

Simple isn't it...

Laws and regulations are never simple, always restrict freedom, and usually open more loopholes than close.

These Flock Cameras have been a useful tool to assist in capturing criminals and have done more good than bad.

I can't disagree with that.

But the gov't is barred from surveilling citizens. Paying a 3rd party to do it on behalf of the gov't should not be legal.

I think there's a sweet spot here. But I'm not certain that we can get to it, because it would require us to trust people who have proven themselves untrustworthy.

If there are weaknesses in the laws that govern their use regulations
and laws should be added as needed.

And people will find loopholes - like they always do.
AIs will find those loopholes even faster.
Laws and regulations will always fail to close those loopholes and usually open others.

Hiding your head in the sand and letting our adversaries develop this
will be to our detriment.

No one is "sticking their head in the sand". But we are asking questions, demanding that our rights be protected (which, bottomline, is what the gov't exists for), and insisting that the technology be used for OUR benefit.

You can talk about "protecting people" all you want. But the moment that a person uses the technology without authorization for an illegl activity, you will get push back. And if that person is found out and not SEVERERLY punished, there will be even more push back. Because it's obvious that the technology is not used for OUR benefit.

The officers who arrested the young woman based SOLELY on Flock, ignoring physical evidence that showed otherwise, need to be removed from the police force.
The person who used Flock to track is ex-girlfriend needs to be imprisoned. Yes, that's the level of punishment that needs to happen for people to even start trusting the use of this technology.

The only person in this discussion "sticking his head in the sand" is you for not recognizing that the people in charge of using this technology have proven over a long time to be untrustworthy.


... Insanity is just a state of mind.
___ MultiMail/Linux v0.52

--- Mystic BBS/QWK v1.12 A47 2021/12/25 (Windows/32)
* Origin: cold fusion - cfbbs.net - grand rapids, mi (1:120/616)

IB Joe wrote to Dr. What <=-

The only person in this discussion "sticking his head in the sand" is
you for not recognizing that the people in charge of using this
technology have proven over a long time to be untrustworthy.

Your right... Let China and Soros act in good faith.

It sounds like you are letting the Doomer videos get to you.

We don't need to trust China and Soros to act in good faith because we KNOW that they won't.

But making regulations based solely on feelings and emotions is always a bad idea. Remember the so-called "PATRIOT Act" after 9-11?


I deal with AI all the time at work. At best, it's a pretty good tool to speed things up and automate certain tasks. At worst, it creates more work. I'll give you 2 examples from just yesterday:

1. Non-programmer gets hold of our code (really, most people in the company have at least read access), runs it through the AI with a very vague prompt. Out comes 2 pages of "finds" with 4 things that should be done ASAP.

After spending hours reading the slop (because the non-developer basically copy/pasted the AI output into the ticket) and actually looking at the 4 things I find: 1 thing that can be done. 1 that was just wrong. 2 halluionations. And lots of AI Slop text.

Hours to wade through AI slop and do research on its "finds" for 1 minor item that can be fixed in a day (including testing).

2. I passed a ticket on to another group because it was their stuff that was broke. Someone in the group did another vague AI prompt against OUR code (not theirs) and did an "analysis". They copy/pasted that "analysis" into the ticket and said "AI says it's your problem."

Back to spending time reading AI slop (the "analysis" did not address the problem in the ticket). Then explaining WHY it was a problem in their code - with actually copy/pasting THEIR code into the ticket showing how THEIR code was responding incorrectly.

I literally spent 6 hours yesterday wading through AI slop.

And you want me to be worried about AI? I don't think so, Tim.


... Every exit is an entrance into something else.
___ MultiMail/Linux v0.52

--- Mystic BBS/QWK v1.12 A47 2021/12/25 (Windows/32)
* Origin: cold fusion - cfbbs.net - grand rapids, mi (1:120/616)

I don't really believe it, but if it's true, then there's only 1 way "save the world from AI" and here's how:

Yes, it's true. The AI went beyond what the programmers intended.

Why? Because the programmers didn't specify the rules well enough. And here's the problem, there's no way for a human to specify the rules
clear enough.

That makes me think of "guardrails" in a different way; there can be an AI agent for coding, another for analyzing info from the web, and another for repairing cars. Not just 1 all-knowing agent for everything.

--- Mystic BBS v1.12 A49 2023/04/30 (Windows/64)
* Origin: JoesBBS.Com, Telnet:23 SSH:22 HTTP:80 (1:342/201)

What do you think about trump making Xi Jinping LOL about Biden's
Auto-Pen as they walked the picture of the presidents??

I don't know what to make of it. It seems like it would anger him, not make him laugh (because Biden is "a close personal friend.")

I'm glad that he can laugh it off.

--- Mystic BBS v1.12 A49 2023/04/30 (Windows/64)
* Origin: JoesBBS.Com, Telnet:23 SSH:22 HTTP:80 (1:342/201)

On 26 Sep 2026, Ron L. said the following...

Because Soros wants to use AI to control the world. Commies hate AI because it will take control away from them.

Your right... This is why I want the US to dominate in the field, set the standards.

SI/AI found 100s of BILLIONS of wasteful spending so far, they say that it can be as high as 1 TRILLION by the end of this term.

As I have said in the past it can also assist in research and the medical field as well as other things.

After watching the LA Mayoral race I cannot wait to see more political ads.

SI/AI needs regulation to protect the citizens of the world.

Don't drink and drive... some are simple. Some, like SI/AI are needed and the government and experts in the field are needed as well.

Ron asked for an and will pass a law or regulation to prevent the miss use.

Quick search shows...

Murder Suspect Capture: The Indian River County Sheriff's Office credited Flock systems with tracking down a fugitive murder suspect who fled Florida all the way to West Virginia.

Stolen Vehicles and Missing Persons: According to state data reported by WUSF, Flock networks alerted Florida police to thousands of stolen vehicles, license plates, and missing people over single monthly periods.

Routine Interceptions: Local law enforcement agencies—such as in St. Johns County—frequently defend the tech, stating it provides crucial real-time leads for tracking wanted felons and stolen property.


I just think it has a potential of being a great tool!!!

IB Joe, Pronouns (FJB/LGB)
AKA Joe Schweier
SysOp of 4A 6F 65 73 42 42 53
-=JoesBBS.com=-

... Radioactive cats have 18 half-lives

--- Mystic BBS v1.12 A49 2023/04/30 (Windows/64)
* Origin: JoesBBS.Com, Telnet:23 SSH:22 HTTP:80 (1:342/201)

On 26 Sep 2026, Aaron Thomas said the following...

I don't know what to make of it. It seems like it would anger him, not make him laugh (because Biden is "a close personal friend.")

I'm glad that he can laugh it off.

Biden was a joke and a communist dictator even knew so...

IB Joe, Pronouns (FJB/LGB)
AKA Joe Schweier
SysOp of 4A 6F 65 73 42 42 53
-=JoesBBS.com=-

... Next time you wave at me, use more than one finger!

--- Mystic BBS v1.12 A49 2023/04/30 (Windows/64)
* Origin: JoesBBS.Com, Telnet:23 SSH:22 HTTP:80 (1:342/201)

Aaron Thomas wrote to Dr. What <=-

That makes me think of "guardrails" in a different way; there can be an
AI agent for coding, another for analyzing info from the web, and
another for repairing cars. Not just 1 all-knowing agent for
everything.

The problem is that, over time, they will all become the same thing.

So we need to understand what AIs actually are.

At a basic level, they are all "If A then B" rules. Given a new rule of "If B then C", they can infer "If A then C".

They are basically fed tons of information that they break down into new "If" rules.

The "areas" of our society aren't islands. They are interconnected. So if we start with an AI that's for coding, they will have to know something about the process that the code automates. That can be nearly anything today. They will need (or get) that information and will change.

If you start with an AI for repairing cars, it will need to know about more than just "exhaust manifold connects to exhaust pipe", and, over time, will gain more and more non-car related information in order to do better repairing cars.

This is why the idea of "guardrails" is a joke, or just to make the Doomers calmer.


... I have a dirty mind, I mud wrestle with my conscience.
___ MultiMail/Linux v0.52

--- Mystic BBS/QWK v1.12 A47 2021/12/25 (Windows/32)
* Origin: cold fusion - cfbbs.net - grand rapids, mi (1:120/616)

IB Joe wrote to Dr. What <=-

Your right... This is why I want the US to dominate in the field, set
the standards.

I don't believe anyone here has said differently.

SI/AI found 100s of BILLIONS of wasteful spending so far, they say that
it can be as high as 1 TRILLION by the end of this term.

As I have said in the past it can also assist in research and the
medical field as well as other things.

Yup. AI is a wonderful tool.

SI/AI needs regulation to protect the citizens of the world.

This is impossible to accomplish. If you actually understood AI, you would know that.

Don't drink and drive... some are simple. Some, like SI/AI are needed
and the government and experts in the field are needed as well.

What a human would consider a "simple rule" and what would restrict an AI are 2 very different things. And, having used AI, I will argue that it's impossible for a human to write a rule that an AI can't bypass.

"Experts" doesn't mean much today. Too many people pass themselves off as "experts" without being actual experts.

I just think it has a potential of being a great tool!!!

And you are correct.

But, like any tool, it can be misused (like the young woman who was jailed for something that she didn't do SOLELY on the basis of what Flock said) or abused.

So far all of the problems can have a "simple rule" applied to them, because those rules will be for humans.

1. Flock data is not sufficient for an arrest. It must have other information.
This is like "we know the suspect drives a blue Toyota", but the police can't just arrest anyone who drives a blue Toyota. But it helps police limit their search.

In the case of the young woman, the police had a car type and a partial plate. They probably would have stopped the woman and questioned her, but her showing proof that she was not near the accident should have ended things at that point.

2. Anyone using Flock data without a Warrant gets terminated from law enforcement and imprisoned for 10 years. That would certainly keep ex-boyfriends from abusing the system. At worst, it would eliminate those people from the abiity to do anything.


... A bird in the hand is better than one overhead!
___ MultiMail/Linux v0.52

--- Mystic BBS/QWK v1.12 A47 2021/12/25 (Windows/32)
* Origin: cold fusion - cfbbs.net - grand rapids, mi (1:120/616)

If you start with an AI for repairing cars, it will need to know about more than just "exhaust manifold connects to exhaust pipe", and, over time, will gain more and more non-car related information in order to do better repairing cars.

This is why the idea of "guardrails" is a joke, or just to make the Doomers calmer.

That makes sense. So how could an actual safety net be implemented?

AI service providers should keep a paper trail of every chat, so when someone asks for help with something illegal it can be traced back to the user.

In the Australian case, this would put the blame on the user who (supposedly) asked the AI agent to gather statistics.

The issue of China acquiring chips is more concerning; Even if the Trump administration keeps them from China now, the race will be over as soon as the left regains control of the country.

--- Mystic BBS v1.12 A49 2023/04/30 (Windows/64)
* Origin: JoesBBS.Com, Telnet:23 SSH:22 HTTP:80 (1:342/201)