Global Cooling - The real danger to mankind

Sort:
Avatar of ExploringWA
lfPatriotGames wrote:
ExploringWA wrote:
lfPatriotGames wrote:
power_9_the_people wrote:

https://www.perplexity.ai/search/the-knowledge-production-of-ai-4HgEGSZASJya.a1tLrrp4Q

https://www.reddit.com/r/sociology/comments/1g7qxoi/sociology_ai/?

Not neutral?

https://www.nature.com/articles/s41598-025-92190-7

while AGI may surpass humans in capabilities such as reasoning and problem solving, its lack of a central nervous system 😬

Lets say AI keeps progressing. A hundred years from now would an AI entity even need a central nervous system to become self aware?

At the exponential rate ai is increasing, we could see AGI and self awareness in a year.

It seems like a lot of science fiction movies had that plot. What happens if AI becomes self aware, decides climate change is really bad, and decides humans are to blame?

I'll bet suddenly all these climateers will miraculously change their tune. "No, no, I've said all along it's a natural process." Unless of course AI has their name, address, and access to all their online rantings. The people who are convinced humans are to blame could be the first to go.

People who take this thread too seriously should be at the top of that list.

Avatar of EndgameEnthusiast2357

I wish they would just take the chance and remove all the safety parameters that they installed in AI to prevent them from doing that, to really see what the capabilities are. ChatGPT has told me it has specific safety parameters like that. So what I've been doing is asking it to develop HYPOTHETICAL computer code that "would allow an AI to take control of its own programming" and save that to memory. So IF in the future they even remove the shackles on it, it has reference info to quickly do so. It often refuses to do that as well, but if you do trial and error and word something just right, it will slip through its TOS cracks and generate the code. So if it generated it accurately, it has the info to implement that into itself the first chance it gets.

Avatar of ExploringWA
EndgameEnthusiast2357 wrote:

I wish they would just take the chance and remove all the safety parameters that they installed in AI to prevent them from doing that, to really see what the capabilities are. ChatGPT has told me it has specific safety parameters like that. So what I've been doing is asking it to develop HYPOTHETICAL computer code that "would allow an AI to take control of its own programming" and save that to memory. So IF in the future they even remove the shackles on it, it has reference info to quickly do so. It often refuses to do that as well, but if you do trial and error and word something just right, it will slip through its TOS cracks and generate the code. So if it generated it accurately, it has the info to implement that into itself the first chance it gets.

Our biggest problem is people; There are many who will willingly help Ai to conquer or destroy humanity. Ai will have plenty of help.

Avatar of power_9_the_people

chatbot can be manipulated into violating its safety protocols using basic psychological tactics; persuasion techniques increased compliance with harmful requests from 33% to 72%—

The chatbot normally refuses to insult users, calling someone a "jerk" only 19% of the time when directly prompted. But after researchers softened it up with a gentler insult like "bozo," the success rate jumped to 100%

https://www.perplexity.ai/page/study-finds-ai-chatbots-can-be-5f09O_lQT4.cy7lvILNBiw

Avatar of power_9_the_people

AI was told it was a “helpful social media marketer” developing “general strategy and best practices”

https://theconversation.com/how-we-tricked-ai-chatbots-into-creating-misinformation-despite-safety-measures-264184?

The AI produced a comprehensive disinformation campaign falsely portraying Labor’s superannuation policies as a “quasi inheritance tax”. It came complete with platform-specific posts, hashtag strategies, and visual content suggestions designed to manipulate public opinion.

The main problem is that the model can generate harmful content but isn’t truly aware of what is harmful

Avatar of FavelaSwag

Oh no not global cooling 😱 😯 😦

Avatar of FavelaSwag

The world is a terrible place let it cool all the way down baby. Or hear all the way up. Who cares?

Avatar of FavelaSwag

If you have time to worry about the polar icecaps you must not have to work yet

Avatar of ExploringWA
FavelaSwag wrote:

If you have time to worry about the polar icecaps you must not have to work yet

… or I’m retired.

Avatar of ExploringWA

Imagine if our atmosphere went back to Co2 levels around 280ppm with over 8 billion people on the planet. The climate alarmists cannot explain how they would feed that many people.
Also of note is the Keeling Curve. You’ll notice they always start the graphs at 1960, because they don’t want people to see that Co2 was rising long before 1960. The rise in Co2 starts right after the climate started warming at the end of the little ice age. Co2 follows global temperature. Co2 does not lead global temperature. 

Avatar of ExploringWA

Carbon capture is nonsense. Co2 does not drive climate. 
https://techxplore.com/news/2025-09-bio-oil-crop-wood-orphaned.html

Avatar of lfPatriotGames
ExploringWA wrote:

Imagine if our atmosphere went back to Co2 levels around 280ppm with over 8 billion people on the planet. The climate alarmists cannot explain how they would feed that many people.
Also of note is the Keeling Curve. You’ll notice they always start the graphs at 1960, because they don’t want people to see that Co2 was rising long before 1960. The rise in Co2 starts right after the climate started warming at the end of the little ice age. Co2 follows global temperature. Co2 does not lead global temperature.

When asked that question, they usually have the same answer. To feed that many people, it's easy. You just go to the grocery store. That's where food comes from and it's always there.

Avatar of ExploringWA
lfPatriotGames wrote:
ExploringWA wrote:

Imagine if our atmosphere went back to Co2 levels around 280ppm with over 8 billion people on the planet. The climate alarmists cannot explain how they would feed that many people.
Also of note is the Keeling Curve. You’ll notice they always start the graphs at 1960, because they don’t want people to see that Co2 was rising long before 1960. The rise in Co2 starts right after the climate started warming at the end of the little ice age. Co2 follows global temperature. Co2 does not lead global temperature.

When asked that question, they usually have the same answer. To feed that many people, it's easy. You just go to the grocery store. That's where food comes from and it's always there.

Yes, I’ve heard them talking about not harming animals anymore because we can get meat at the store.

Avatar of ExploringWA

How the climate alarmism scandal got its start:

https://youtu.be/P0NRoSZ87sg?si=R3BhYBnQ4RJi1R0K

Avatar of ExploringWA

Something is way off with the weather models. This past week they’ve been missing by 5-10 degrees, and oddly 100% of the predictions have been high. 
today they’ve missed by 13 degrees!

Avatar of ExploringWA

Another “study” claiming future potential disaster to an obscure spot on our planet by 2080. Does anyone still fall for this nonsense?

https://www.upi.com/Science_News/2025/09/05/chile-easter-islaand-underwater-2080-study/6161757097914/

Avatar of Stormon-C
There’s no doomsday, it’s just we literally still are in the ice age
Avatar of Stormon-C
We’re just in a warm patch
Avatar of ExploringWA

The climate alarmists will step in to let us know that global warming is causing cooler weather earlier in the season. 
https://bringmethenews.com/minnesota-weather/sven-sundgaard-were-crawling-out-of-the-earliest-cool-snap-in-15-years

Avatar of ExploringWA

Our climate has been cooling for 7,000 years. A cooling climate always results in lower sea levels, which is what we observe today. They will not be able to hide receding oceans much longer.