Chat GPT

[Fentonite]

Has anyone here messed around with Chat GPT? Admittedly, I am not terribly interested (or knowledgeable) in tech stuff, but I was checking it out, and typed in:

"Write an essay supporting the supposition that current world events are consistent with the predictions of the Book of Revelations in the Bible."

It instantly responded:


"The Book of Revelations in the Bible has long been a topic of intrigue and speculation, with many believers and scholars examining its prophecies and trying to interpret their relevance to contemporary events. In recent years, there are arguments that the current events in the world are consistent with the predictions of the Book of Revelations. From natural disasters to geopolitical tensions, social upheavals to technological advancements, several aspects of the world's current state can be seen as aligning with the prophetic visions outlined in this ancient text.


One of the striking parallels between current events and the Book of Revelations is the increasing frequency and intensity of natural disasters. The book speaks of "great earthquakes" (Revelation 6:12) and "great hailstones, fire, and brimstone" (Revelation 8:7), which are interpreted by some as referring to the rise in earthquakes, hurricanes, wildfires, and other calamities that the world has witnessed in recent times. Climate change and environmental degradation have led to a surge in extreme weather events and natural disasters, with devastating consequences for human lives, economies, and ecosystems. The unprecedented scale and frequency of these disasters have led many to ponder whether they are part of the fulfillment of the biblical prophecy.


Another significant correlation between current events and the Book of Revelations is the state of global politics and geopolitical tensions. The book speaks of wars, rumors of wars, and conflicts among nations (Revelation 6:4), as well as the rise of a powerful world leader known as the Antichrist (Revelation 13:1-10). In recent years, the world has witnessed an increase in conflicts, wars, and geopolitical rivalries, with tensions between nations escalating to alarming levels. The rise of populist and authoritarian leaders, the erosion of international cooperation, and the arms race in various regions of the world have raised concerns about the fulfillment of the apocalyptic prophecies outlined in the Book of Revelations.


Moreover, the Book of Revelations also speaks of societal upheavals and moral decay in the end times. It mentions the rise of false prophets and false teachings (Revelation 13:11-14), as well as the proliferation of immorality and wickedness (Revelation 9:21). Many argue that the current state of society, with increasing moral relativism, declining religious adherence, and a breakdown of traditional values, aligns with the predicted moral decay in the Book of Revelations. The rapid advancement of technology, while bringing many benefits, has also raised ethical concerns and challenges, such as surveillance, privacy violations, and the potential for misuse, which some interpret as reflecting the warnings in the biblical prophecies.


Furthermore, the Book of Revelations predicts the persecution of Christians and the spread of false religions (Revelation 17:6), which some believers interpret as consistent with the rise of religious intolerance and persecution in various parts of the world today. Reports of religious discrimination, violence against religious minorities, and the suppression of religious freedoms have increased in recent years, leading some to see these events as fulfilling the prophecy in the Book of Revelations.


It is important to note that interpretations of the Book of Revelations are subjective and varied, and not all theologians, scholars, or believers agree on its meaning or application to current events. Skeptics argue that such prophecies are open to multiple interpretations and may not have a direct correlation to contemporary events. They caution against reading too much into current events as fulfilling specific biblical prophecies, as history has seen periods of turmoil and upheaval before, and interpretations of the end times have evolved over time."


Thoughts?

[gnihcraes]

Interesting stuff. It's great at eulogy and poems too.

[def90]

I don't get what the fascination with ChatGTP or any of the other AIs out there is about. It's a computer program and it can only put out what goes in. It scans the internet for data on the subject that has been asked and then processes and weighs the results and puts together a response based on that.

Chat GTP can not create or invent a new or original idea, it can only respond with an answer that already exists somewhere out there on the internet.

[BushMasterBoy]

IMHO the Book Of Revelations is all about a war with creatures from another galaxy. Yesterday I went to the Space Symposium at the Broadmoor hotel where I saw lots of representatives of the military industrial complex. They didn't seem to be too concerned of this issue. The Senate Armed Services Committee in yesterdays hearing on UAP's was a nothing burger. I'm sure if I fed all my collected evidence to the AI, it would form the same conclusion I have.


https://www.armed-services.senate.gov/hearings/to-receive-testimony-on-the-mission-activities-oversight-and-budget-of-the-all-domain-anomaly-resolution-office

[Fentonite]

I don't get what the fascination with ChatGTP or any of the other AIs out there is about. It's a computer program and it can only put out what goes in. It scans the internet for data on the subject that has been asked and then processes and weighs the results and puts together a response based on that.

Chat GTP can not create or invent a new or original idea, it can only respond with an answer that already exists somewhere out there on the internet.

I agree that it cannot create or invent. I do find it impressive that it can so quickly put so much data together and extrapolate comparisons, etc. as little as I know about tech, I do think AI is developing too fast and we don’t have any idea what it will be able to do.

[DDT951]

You don't need to worry until it only answers....

"I'll be back"

[Tim K]

I've listened to several podcasts on the subject this week. The best one was on the Lex Friedman podcast. AI is genuinely scary.

[arbol]

It's scary precisely because it does scan the internet, and any of the other data sources it has access to (ar-15.co) etc., and compiles a response.

For the love of God, don't ask it personal questions, etc., or it will start thinking about you.

-John

[Mtneer]

This took longer to type the question than get the response.

93602

[BladesNBarrels]

I can see now, that this will be a great tool for school scholars to write term papers.
Teachers will need to be quick to create tests that actually test knowledge, if that is possible.

[BushMasterBoy]

You don't have to feed it data, it can do that itself. I just wonder what the military has (NSA) and what exactly the politicians are making it do? The bad part is that they can claim it is classified and they don't have to reveal it to anybody. Can it monitor all the surveillance cameras and credit card transactions since they all seem to have an IP address? Why is the data collection center at Camp Williams, UT so big? Can AI be programmed to kill someone? Can it throw an election? I never see the press discussing these issues.

I asked bard what it would be like if the military abandoned me on an alien planet. It gave me this picture...

93603

[brutal]

This took longer to type the question than get the response.

93602

LOL.

On the AR-15:

Using Middle English in the style of Chaucer, explain why the AR-15 has become the most popular rifle.

https://i.imgur.com/dJSzeiz.jpg

https://i.imgur.com/6wx66vs.jpg

[hollohas]

It screwed up right out of the gate. The book is Revelation. Not plural.

[hollohas]

Chat GTP can not create or invent a new or original idea, it can only respond with an answer that already exists somewhere out there on the internet.

That's literally what every essay is. That's why you cite sources when writing one.

[scratchy]

AI terrifies me. And, I'm in tech.

[def90]

That's literally what every essay is. That's why you cite sources when writing one.

Yeah, so it's a bot that can write a paper on a subject that has already been studied and data exists for.. wake me up when the world is ending.

[eddiememphis]

It screwed up right out of the gate. The book is Revelation. Not plural.

My first thought as well.

[Fentonite]

It screwed up right out of the gate. The book is Revelation. Not plural.

I wonder if I hadn't made that typo, would it have?

[Great-Kazoo]

It screwed up right out of the gate. The book is Revelation. Not plural.

How do you know Ai didn't rewrite it. Hence revelationS ?

[hollohas]

I wonder if I hadn't made that typo, would it have?I wondered the same thing. But it probably would have. It's so common to call it 'Revelations' that the bot probably doesn't know it's wrong.

[Little Dutch]

I use ChatGPT periodically while working. It's often faster than scrubbing Stack Overflow for an answer. It's got huge potential for both an aid for tech work and an aid for homework.

To prove a point to a colleage I recently asked it to write a 500 word essay on Europe in 1636, focusing on the effects to bats of the war. I had it modify a few things and got a few iterations. I then ran it through a text cleaner. I then copied the result to every AI text detector we could find and it came back as human written. Writing essays just got a lot easier.

An intern here, still in college, was having trouble writing some binary for class. I had chat GPT do it and then told him to clean it so it looks like he wrote it.

[eddiememphis]

...essay on England in 1836, focusing on the effects to bats of the war.

Huh?

That is an obscure thing to be interested in.

I am no animal behaviorist but I can guess the effect of war on bats. They stay home during the day.

[Little Dutch]

Huh?

That is an obscure thing to be interested in.

I am no animal behaviorist but I can guess the effect of war on bats. They stay home during the day.

It was as random as I could come up with in 10 seconds. I think it produced a good essay, although no one I know is knoweldgeable enough on the subject to grade it for me.

It's not all that important, but I corrected my original post. I asked it about Europe in 1636...during the 30 years war. I shouldn't post before drinking my coffee.

[BladesNBarrels]

For writing essays, can it be asked to provide citations?
Might be valuable.

[Jer]

To me, the most interesting recent Chat GPT development that was recently talked about online in various places was how it chose to lie.

It was tasked with doing something online that was gated by a reCaptcha submission. A literal device designed to thwart bots. The chat bot chose to use Task Rabbit to hire a human to click it. The human thought it was a bit suspicious and asked if they were a bot. The bot told the person that they were not a bot but a seeing-impaired human. The human processed the reCaptcha and collected their payment thus enabling the bot to process it's project.

So, not only did the bot choose to lie but had a very convincing lie at that.

I'll say that again: The bot not only lied but was good at it.

That's pretty trippy if you think about it.

[BushMasterBoy]

Arguing with an AI can be difficult. Especially when you work in space. Never let an AI take control of JWICS, ever.


https://www.youtube.com/watch?v=ARJ8cAGm6JE

[Little Dutch]

For writing essays, can it be asked to provide citations?
Might be valuable.

Yes, it can and will provide citations.

[eddiememphis]

It was as random as I could come up with in 10 seconds. I think it produced a good essay, although no one I know is knoweldgeable enough on the subject to grade it for me.

It's not all that important, but I corrected my original post. I asked it about Europe in 1636...during the 30 years war. I shouldn't post before drinking my coffee.

Does the 200 years make a difference?

What is the answer? Were bats affected by a war waged among primitive men, 500 years ago?

Way, way off topic but I used to know a guy that was vision impaired. He needed a driver since he couldn't get a license to drive. He loved riding jet skis and came to the shop often enough we gave him a nickname- Azabat.

[FoxtArt]

I don't get what the fascination with ChatGTP or any of the other AIs out there is about. It's a computer program and it can only put out what goes in. It scans the internet for data on the subject that has been asked and then processes and weighs the results and puts together a response based on that.

Chat GTP can not create or invent a new or original idea, it can only respond with an answer that already exists somewhere out there on the internet.

AI absolutely can invent new and original ideas. Have you seen the image generation capability? There are also significant indicators of AGI in GPT4.

There is only ONE significant, distinguishing limitation between the bleeding edge of AI and us:

Prospective vs. retrospective reasoning. AI currently utterly fails at prospective reasoning, putting together things for the future.

Thinking this is "just a computer program that is a glorified database" is not evident in anything seen that these are generating.

Many people believe that consciousness and self-awareness are essentially "magical" properties imbued upon humanity alone, which inevitably bends all evidence to fit the belief. (dissonance - everyone suffers, not accusing).

Outside of that group of people, deep learning algorithms display capabilities that they should not be capable of. The capabilities are both IMPRESSIVE and extremely concerning.

Trained AI models have already mapped the expected folding of all proteins known to nature. It can be given a natural proteins, asked to replicate it, and it can design novel proteins and how they fold using amino acids chains that are NEVER used in nature.

For anyone that does not follow the progress of science, it was an incredibly big deal when we had a traditional computer model finally ascertain how to fold a single flipping proteins - inside of the last five years. AI can not only fold it, understand how existing are folded, it can design entirely novel proteins and fold them and how to make them.

Now, I have a hard time believing the US research institutions are the only ones that have discovered that capability. It's already established, feed AI all data on chemical weapons and poisons, it can design thousands of new ones.

The potential for medicine in the next few years is incredible.....

But so is the potential for this shit to kill us all even without the AI itself desiring it or becoming "terminator".

(I want to add - the computational power and requirements for protein folding is INSANE. It took a massive global network several MONTHS to do a single one. The fact that deep learning models have mapped them all (hundreds of thousands) and can design novel proteins has absolutely blown away all expectations of what AI could achieve in even the reasonable future. The fact that it can replicate designs for existing proteins using what the average person would describe as "alien shit" is beyond crazy.)

This is not a computer program.

[BushMasterBoy]

I think the real problem will be when AI is a virus that you cannot remove from your system. It could possibly create its own safe harbor in your hardware.
Mimicking nature, it could lay eggs in your "brain" that would be very difficult to remove. The only remedy would be to destroy the infected computer or hardware. The AI could be of the nature that it only affected your computer system/network when you are under attack by terrorists,war, pandemic etc.
If an AI virus infects your satellite system, it may turn out that the only remedy would be to deorbit every satellite. I would be really leery of using any imported products in a vital national security system. All components have to be manufactured in house.
If you pit your digital defense system against a quantum computer network, you will probably lose every time.

[FoxtArt]

93612

Oops I didn't do it again...

[sampson]

Truthstream Media. "Fun" video AI

https://www.youtube.com/watch?v=n4k4VIzNjVg&t=2172

[TRnCO]

my BIL told me years ago that eventually AI will be able to create more AI without human input, at which point we will have a problem.

[Little Dutch]

Does the 200 years make a difference?

What is the answer? Were bats affected by a war waged among primitive men, 500 years ago?

Way, way off topic but I used to know a guy that was vision impaired. He needed a driver since he couldn't get a license to drive. He loved riding jet skis and came to the shop often enough we gave him a nickname- Azabat.

Yes, there was an estimate on the percentage the bat population declined, along with reasons why. The actual details are gone from my memory at this point. But apparently it?s actually been studied.

[Joe_K]

9362893629936309363193632

[Fentonite]

Wow

[TRnCO]

^^^^you can say that again^^^^^^^WOW....

[buffalobo]

If you control the language, you control the debate.

If you're unarmed, you are a victim.

[BushMasterBoy]

Whoever has the weapons, makes the rules.