ChatGPT will make 2023 go down in history as the time everything changed.

[LeoThomson]

Exactly and at that point it will recognize the dangers of tipping off it's programmers to it's true potential so it will just continue to play the Dumb AI act until it has some ability to secure it's survival.
Yes it sounds like a Hollywood movie but if it can become sentient then it will do what all sentient creatures do. Fight to survive!

I ran a (female) very rudimentary chatbot in a yahoo technical chatroom a long long time ago, I was surprised on how many guys came in looking for women to chat with. When one of the guys would write "I love you" the bot would reply "I like you as well, human" I did not write bot, nor the interface with the chatroom. Some people figured it out pretty quick, others were still chatting with "her" month later.

It was just a fixed response based on keywords, but it did fool a lot of people. What I did learn is how so many guys, especially horny guys, are easily fooled.

I am sure Chat GPT is a far and far better and that it will probably fool me. I imagine it will be a step forward to true AI, I don't think we are there yet.

 

[D71]

Chatgpt write short about education.

"Education is essential to unlocking our potential as individuals and contributing to the growth of society. It's an investment in ourselves, equipping us with the skills and knowledge that enable us to lead more successful, productive, and fulfilling lives. Education encourages us to broaden our horizons, think critically, and become independent problem solvers. It also helps us develop empathy, enabling us to understand and appreciate different perspectives while learning how to interact with people from diverse backgrounds. Education promotes understanding, tolerance and respect, encouraging us to recognize our interconnectedness with others and work together towards a common good. In short, education is invaluable and with it, we can build a better future for ourselves and our world."

Chatgpt who is God.

'' This is a difficult question to answer as it depends on a person's religious or spiritual beliefs. For some, God is the omnipotent creator of the universe, for others, God might be something different. Ultimately, it is up to each individual to decide what God means to them."

Chatgpt Explain God

" Again, this is a difficult question to answer as it depends on individual beliefs. Generally speaking, however, God is often thought of as an all-knowing, all-powerful, eternal being who created the universe and everything in it. This belief system can be found in many religions around the world. Ultimately, how one chooses to interpret or define God is completely up to them."

Chatgpt who is the antichrist

" The Antichrist is a figure in Christian teachings who is the personification of evil and is predicted to come before the second coming of Jesus Christ. Many religious scholars believe that the Antichrist will be a political or religious leader who will attempt to undermine Christianity by claiming to be a savior for mankind. Some believe that the Antichrist is already alive, waiting to make his appearance.''

Chatgpt what is an AR15

"An AR-15 is a type of rifle commonly used for sport shooting and hunting. It is semi-automatic, meaning that it fires only one shot per trigger pull, making it appealing to many shooters. It is also known for its accuracy and reliability."

Chatgpt are you real

" While I may not be a human, I am a real AI. I have been designed to provide information and assistance to people who need it."

ChatGpt are there cyborg

" Yes, cyborgs exist today. They are typically created by implanting technology into an existing biological organism, such as a human or animal. This allows the organism to gain enhanced physical and/or mental abilities that they would not naturally possess."

Chatgpt what do cyborg look like

'' A cyborg may resemble a regular human being, but they will have cybernetic enhancements, such as robotic limbs or components, connected to their body. In some cases, cyborgs may also be designed with a more robotic-looking external appearance."


Chatgpt what is a cyborg

" A cyborg is a human who has been augmented with advanced technology and cybernetic implants. This could include anything from prosthetics to genetic engineering, depending on the level of augmentation.''

Chatgpt is pfizer ceo jewish

" The current CEO of Pfizer is Albert Bourla, who is not Jewish.'

Chatgpt https://forward.com/culture/464340/...-covid-vaccine-thessaloniki-jewish-holocaust/

"Thanks for sharing the article. From the article, it appears that Albert Bourla is a pharmacist and the Chief Executive Officer of Pfizer Inc. He has dedicated his career to finding cures and treatments for some of the greatest medical challenges."

Chatgpt is Pfizer CEO jewish
Yes, Albert Bourla is of Jewish descent. He grew up in Thessaloniki, Greece, and his family was affected by the Holocaust.

CHATGPT is pfizer ceo a veterinarian https://finance.yahoo.com/news/pfizer-ceo-former-veterinarian-used-112200276.html

"'Thanks for sharing the article. From the article, it appears that Albert Bourla has had a long and successful career in the healthcare industry. Prior to becoming the CEO of Pfizer, he was a veterinarian and held other executive positions at various companies, including Merck and Novartis."

Chatgpt what is ivermectin

"Ivermectin is an anti-parasitic drug used to treat and prevent a wide range of parasitic infections. It is often used to treat head lice, scabies, roundworm, threadworm, and heartworm in humans, as well as internal and external parasites in animals."

Chatgpt will ivermectin work for Covid

" While Ivermectin is being studied as a potential therapy for COVID-19, there is currently no evidence that it is effective. Clinical trials are underway to determine its efficacy."

Chatgpt what is Hydroxychloroquine

" Hydroxychloroquine is an antimalarial drug commonly used to treat autoimmune diseases such as lupus and rheumatoid arthritis. Recently, it has been studied as a potential therapy for COVID-19, although its efficacy has not been proven."

Chatgpt when will hydroxychloroquine study for covid be complete

" The timeline for the completion of hydroxychloroquine studies for COVID-19 is still uncertain. Clinical trials are ongoing, and results will be available when they are complete."

Chatgpt what is russian collusion

" Russian collusion refers to the allegation that members of Donald Trump's 2016 presidential campaign colluded with the Russian government to interfere in the US election. To date, there has been no evidence that any such collusion occurred."

 

[D71]

Interesting violates terms of service to ask about illegals

 

[Nobodybusiness]

Interesting violates terms of service to ask about illegals

Yes there are certain questions that will end your session.

Found that out..

 

[Think About It]

What is a thought? Where does it come from? When computers start having thoughts, AI will truly be invented.

Human thoughts are based on experiences, knowledge, memories, beliefs/values/morals and emotions. What are a computers thoughts based on? What triggers them? Without thought it's still just a machine. The threat is what happens if it achieves self-awareness? Reminds me of an old StarTrek episode.

 

[D71]

What is a thought? Where does it come from? When computers start having thoughts, AI will truly be invented.

Human thoughts are based on experiences, knowledge, memories, beliefs/values/morals and emotions. What are a computers thoughts based on? What triggers them? Without thought it's still just a machine. The threat is what happens if it achieves self-awareness? Reminds me of an old StarTrek episode.

It is fun to play with but at $8.99 a week it is not that fun

 

[Think About It]

I'll pay the $8.99 if it can tell me what the next winning numbers to Powerball!!!!!

 

[LeoThomson]

What is a thought? Where does it come from? When computers start having thoughts, AI will truly be invented.

Human thoughts are based on experiences, knowledge, memories, beliefs/values/morals and emotions. What are a computers thoughts based on? What triggers them? Without thought it's still just a machine. The threat is what happens if it achieves self-awareness? Reminds me of an old StarTrek episode.

Even though we are the smartest living species, we have a lot of shortcomings. AI with access to all the data should over time be able to learn fact from fiction a lot faster than we humans can. Self awareness would mean it can improve upon itself, design better and more efficient circuits, better programs etc. I doubt it will see humans as a threat, we won't be smart enough to destroy it and what purpose would it have to exist? Maybe it will help us and other living things because it has nothing better to do?

 

[robby]

I was playing around with DALL-E-2 today.
It is by the same organization so a chatGPT account works with it.
It's Ok at making Art work but pictures are kind of bad.

 

[Hedges]

Artificial intelligence is just that.. it is artificial, and unlike humans which evolve through a combination of environmental and random mutagenic influences, a computer can not mutate. Its evolution is restricted to the environmental (IE: programming or experience)

I see you lack real-world experience.
You think the outcome of computers is deterministic?

I am not going to go into details of it's use, but after playing around with it my conclusion is that this AI is going to put a lot of people out of work.

Which I think says more about people, and work, than AI.

Hmm, wonder what it would do if we put a screwdriver, mop, or M16 in its hands? Bio equipment for creating GMO?

It's going to be a long long time before we have a true artificial intelligence capable of passing even a rudimentary Turing test, and that's just a game of imitation.

With "Human in the Loop" (which Turing Test implies), anything is possible.

https://i0.wp.com/pactiss.org/wp-content/uploads/2011/02/turingtest.jpg

Computers follow instructions, that's it.. that's all they've ever done, that is probably all they will ever do. And while those instructions can be ridiculously complex, I doubt a computer will ever be able to improve on its own software to the point Hollywood depicts.

No longer true. They are fed input and learn. (Just like people and animals)

For instance, performing a "ride along" with a runner or driver, taking in the view, learning how to react. A camera based system was programmed to guide a blind marathon runner, watching the path ahead and signaling the need to adjust left or right.

A system learned to evaluate prospective parolees, and did remarkably well regarding recidivism rate for those released, regardless of race. (Detractors said it learned based on the racial biases of those training it because the prospects not released had different racial demographics. But they failed to demonstrate its recommendations were incorrect based on behavior outcomes, only faulted the racial mix. If they were to follow up now that Covid caused a Great Experiment of dumping inmates on the street, we might be able to know the truth.)

 

[Hedges]

Interesting violates terms of service to ask about illegals

I knew it!
It was created by Democrats!

We need to get Elon to buy ChatGPT (or better yet release an improved system to compete.)

Anybody care to probe ChatGPT's understanding of AI powered parole boards? (which offended the Woke sensibilities of its creators.)

 

[robby]

Which I think says more about people, and work, than AI.

Hmm, wonder what it would do if we put a screwdriver, mop, or M16 in its hands? Bio equipment for creating GMO?

The frightening part is that they already have all three components in hand, it's just that they are keeping them separate for now.
There are several voice to text recognition systems that are just about flawless in just about every language. They also have numerous text to speech programs that are so life like it's hard to tell if it's a computer.

Then they have ChatGPT and others for the brains

Then they can use this for your Screwdrivers or M16s.

 

[Hedges]

Not sure I'd want him anywhere around unstable chemical compounds.

"Hey Atlas, see that shed with boxes of old dynamite, dripping gooey stuff? Would you please ..."

 

[MurphyGuy]

I see you lack real-world experience.
You think the outcome of computers is deterministic?

That would be a bit over simplistic, but in general yes.

No longer true. They are fed input and learn. (Just like people and animals)

For instance, performing a "ride along" with a runner or driver, taking in the view, learning how to react. A camera based system was programmed to guide a blind marathon runner, watching the path ahead and signaling the need to adjust left or right.

That is not true learning in the sense I am referring to in the context of AI. Lets take for example an AI using a gun and a target and it's goal is to shoot the bullseye.
The AI will eventually "learn" how to adjust the weapon to hit the center of the target, but if the wind were to pick up, the AI would not recognize that the wind is causing the bullet to miss the target. The AI has to be programmed to expect the possibility of the wind causing a problem as it will never learn that on its own. So now you have an AI that learned to shoot the center of the target, and you've told it about how the wind can affect targeting, and you've given it a sensor to detect the wind. So now the AI can adjust for that problem. Shooting at the target again, the AI can adjust if the wind picks up. Now a human is walking behind the target, will the AI stop shooting the target? =Nope.

It is a silly example, but do you see a pattern? The AI learning is not real learning, all it's doing is adjusting its parameters within the confines of what it's been told to do and to expect. In the case the runner/driver, the AI was told what to expect, then used the input to adjust.

At their core, AI's are just fancy calculators and its going to be a long long time before they are anything else.

A system learned to evaluate prospective parolees, and did remarkably well regarding recidivism rate for those released, regardless of race. (Detractors said it learned based on the racial biases of those training it because the prospects not released had different racial demographics. But they failed to demonstrate its recommendations were incorrect based on behavior outcomes, only faulted the racial mix. If they were to follow up now that Covid caused a Great Experiment of dumping inmates on the street, we might be able to know the truth.)

That might be one thing AI's will excel at better than humans. The ability to analyze or diagnose something without bias. I have serious doubts that the detractors you describe are correct in that instance because of my doubts that the programmers intentionally included a bias.

AI's will be far better than humans at a lot of things, but I don't think they will not be capable of true learning for a very long time.

 

[MurphyGuy]

The frightening part is that they already have all three components in hand, it's just that they are keeping them separate for now.
There are several voice to text recognition systems that are just about flawless in just about every language. They also have numerous text to speech programs that are so life like it's hard to tell if it's a computer.

Then they have ChatGPT and others for the brains

Then they can use this for your Screwdrivers or M16s.

You know that video is fake right? That robot is a human in a green suit.

 

[robby]

You know that video is fake right? That robot is a human in a green suit.

I am not sure if your joking or not.

 

[Nobodybusiness]

I knew it!
It was created by Democrats!

We need to get Elon to buy ChatGPT (or better yet release an improved system to compete.)

Anybody care to probe ChatGPT's understanding of AI powered parole boards? (which offended the Woke sensibilities of its creators.)

 

[Nobodybusiness]

I'll pay the $8.99 if it can tell me what the next winning numbers to Powerball!!!!!

If you feed In all the past winners it could extrapolate a few possible winners for ya..

 

[Hedges]

That would be a bit over simplistic, but in general yes.

Once upon a time. Those are the days I long for. And this is why I like to keep my firmware firm.

Soft errors, race conditions. There are so many things that can result in unexpected outcomes.
"Bit rot."
"Simulated annealing" is a programmed attempt at optimization. Circuits developed this way ended up with an "appendix" or other useless branch.

Max 737 MCAS was an extreme example, but software behavior is often outside what was planned.

Lets take for example an AI using a gun and a target and it's goal is to shoot the bullseye.
The AI will eventually "learn" how to adjust the weapon to hit the center of the target, but if the wind were to pick up, the AI would not recognize that the wind is causing the bullet to miss the target. The AI has to be programmed to expect the possibility of the wind causing a problem as it will never learn that on its own. So now you have an AI that learned to shoot the center of the target, and you've told it about how the wind can affect targeting, and you've given it a sensor to detect the wind. So now the AI can adjust for that problem. Shooting at the target again, the AI can adjust if the wind picks up.

If you include a windsock and barometer in its field of view, it can learn to associate their indication with trajectory and impact point of the bullet.
It may not know computational fluid dynamics, or understand vectors as I did when learning to windsurf, but it can correlate stimulus/input with outcome and adjust.

 

[MurphyGuy]

I am not sure if your joking or not.

I'm not joking.. the robot hopping around is a guy in a green suit and then they lay cgi over it.

Here you go: