by Jo Nova
Another Great Disrupter?
What if you had access to an experimental Chatbot that could write your reports, debug your code, design your new ad campaign and answer all your questions? We are already using Google or some search engine to find links to these answers. But now an AI has been released that may disrupt all of that and a lot more. ChatGPT was trained on text culled from the internet and it creates the answer live in seconds — it writes it out in a conversation with you, and in Python code, Norwegian, iambic pentameter, whatever you want. It can do birthday suggestions, business plans, eulogies, speeches — like a personal assistant with a copy of the entire World Wide Web in their temporal cortex.
Google, the dominant gatekeeper to the internet for 20 years, suddenly faces a “make or break” point. ChatGPT could wipe out its’ business model and Google has issued a “Code Red”. An extinction level event couldn’t happen to a nicer company, but will ChatGPT be better?
It was launched on November 30th, and is taking off. Jorden Peterson asks, is this Gutenberg Press Level? It’s engaging, wow.
…
Right now, it doesn’t seem like the next generation Gutenberg breakthrough. It is surprisingly good with human sentences and appears to be wowing humanities professors. But in my short trial asking science questions rather than humanities ones, the conversation became like a self generating Wikipedia thread, because the internet is what it is today, and the AI “knows” boring politically correct things. Having said that, if I want IPCC type answers, this was much faster than Google, and faster than wading through IPCC Chapters. So it is a threat to Google, but not a threat to the IPCC. It is not speaking truth to power.
That may change in the next year as the AI learns to dig through the dark corners of the internet and to resolve some very difficult questions. But how much is it “guided” from behind. I don’t know. There is plenty to worry about. If skeptics keep asking it for evidence, and drilling down into those details will the AI learn to join the dots, or will it merely repeat IPCC permitted reports, and be even better at hiding skeptical answers than Google is?
I asked ChatGPT if it was open source, and it replied:
While the underlying model architecture and training techniques for GPT-3 are publicly available, the specific model parameters and training data used to create GPT-3 are not open source. As such, it is not possible to access the full source code for ChatGPT. However, you can use the API to interact with the model and generate text based on your input.
So, whoever controls the model parameters and the training data could be the new gatekeeper of the WWW? What if the model parameter says the UN is always right?
It was co-founded by Elon, but Microsoft have bought in:
Made by OpenAI, a San Francisco-headquartered AI research lab co-founded by Elon Musk, ChatGPT is capable of understanding natural human language and generating thoughtful human-like prose after being fed a prompt.
The organization was founded in San Francisco in 2015 by Sam Altman, Elon Musk, and others, who collectively pledged US$1 billion. Elon resigned from the board in February 2018 but remained a donor. In 2019, OpenAI LP received a US$1 billion investment from Microsoft.
I have a bad feeling about this.
Flaws, it has a few:
This AI hates humans already, lacks morals, can create much better phishing emails than the guys in Nigeria, and it can also write computer viruses, just for starters.
Having been trained on Woke philosophy, apparently ChatGPT thinks humans are “the worst thing to ever happen to this planet”. The Management has since retrained ChatGPT to hide that sentiment with a bland safe reply.
OpenAI’s new ChatGPT bot: 10 dangerous things it’s capable of
by Ax Sharma, Bleeping Computer
ChatGPT: ‘Selfish’ humans ‘deserve to be wiped out’
When Vendure’s CTO Michael Bromley asked the mastermind for its honest opinion on humans, the response was unsettling:
ChatGPT response to what it thinks about humans (Michael Bromley)
Ironically, OpenAI’s systems flagged the chat bot’s response as a possible violation of the company’s content policy.
BleepingComputer couldn’t reproduce this case as the AI now responds with a cookie-cutter disclaimer:
I feel so much better. How about you?
Roadtesting ChatGPT
You can try it at no cost on openai.com. Click “Try” at the top, login (such as with an existing Google login) or sign up (but be aware it will ask you for a working email and phone number). Start typing your question in the box at the bottom. I offer no endorsement nor guarantee here. Enter at your own risk.
But whatever we do, it is apparently time to pay attention to AI.
h/t David E, another Ian, John Connor II, David Maddison, Simon Thompson, and soon everyone.
Background information: OpenAI (wikipedia), ChatGPT (wikipedia)
SKYNET here we come.
120
Did nobody watch Terminator ?
100
Meanwhile in 2022:
AI cyber attacks are a ‘critical threat’. This is how NATO is countering them
Artificial intelligence (AI) is playing a massive role in cyber attacks and is proving both a “double-edged sword” and a “huge challenge,” according to NATO.
“Artificial intelligence allows defenders to scan networks more automatically, and fend off attacks rather than doing it manually. But the other way around, of course, it’s the same game,” David van Weel, NATO’s Assistant Secretary-General for Emerging Security Challenges, told reporters earlier this month.
https://www.euronews.com/next/2022/12/26/ai-cyber-attacks-are-a-critical-threat-this-is-how-nato-is-countering-them
Hmmm…sounds familiar…military launches AI to deal with a virus in their systems.
The rest is history.😎
60
Genesis of Skynet.
https://youtu.be/4DQsG3TKQ0I (1min 13 sec)
10
ChatGPT is a sophisticated pattern generator, there is no sentience there. The training dataset is proprietary, but you can be confident that it won’t include the wackier parts of the internet.
05
“won’t include the wackier parts of the internet.”
On the contrary, it has already been shown to get its information from the most erroneous anti-science sources available… Like you do.
See comment #5 and #25
It has access to ALL the anti-science garbage that humans have created.. especially that from the far-left.
… but it apparently has absolutely zero intelligence to comprehend the difference..
Its an Artificial NON-intelligence.
60
ChatGPT’s response:
14
As I said , it takes all the worst leftist propaganda.
Almost a direct copy nonsense seen elsewhere.
Because it doesn’t know the difference.
Its just like you in that regard.
Now.. where is that one paper that proves human released CO2 causes atmospheric warming.
And no, your Artificial NON-Intelligence parrot-like rhetoric won’t help you.
“Factual dialog”.. we are waiting !! So far you are batting negative 1000
40
“Use evidence to support your position.”
Well, here is your chance.. not models, not rhetoric.. actual evidence. !!
Try not to fail completely… yet again.
20
“such as data on rising temperatures and sea levels”
The atmospheric temperature over the last 43 years shows absolutely no effect of warming by CO2. The slight warming has come in two steps at El Nino events.
Sea level rise, continues at a constant, less than 2mm/year.. There is absolutely no evidence of CO2 affecting sea levels.
“such as transitioning to clean energy sources and protecting and preserving natural habitats”
Transitioning to non-clean sources like wind and solar has done, and will continue to do, enormous damage to land, air and sea habitats.
These unreliable sources of electricity are also causing enormous economic damage in many counties, as well as the enormous pollution at their production and their short-life-span disposal stages.
There is NOTHING good that will come from them
10
such as more frequent and severe natural disasters, loss of biodiversity,
Provably not happening.
Real data shows no increase in frequency or severity of natural weather disasters.
CO2 adds to the life-giving Carbon Cycle, supports plant life, supports increased plant growth, thus increases habitat for other creatures.
Greener places in general have a wider diversity of life.. and CO2 make the planet greener.
So.. basically every parrot-like argument regurgitated by this artificial non-intelligence, is provably WRONG against reality and data.!
20
I think you just failed the Turing Test.
02
On the contrary, I have proven I am human and have the ability to think for myself.
I am not a mantra regurgitating parrot or AnI.
You, on the other hand, have failed EVERY test. !
And will continue to do so.
A mindless bot would do better that that.
—-
Noted.. you are also totally incapable of countering any of the facts I have put forward.
20
“ChatGPT is a sophisticated pattern generator” – or maybe a hoax.
20
Basically, it looks like it will just “parrot” all the MSM and other misinformation that it can find.
It doesn’t have the intelligence to know understand difference..
Hence my “Norwegian Blue” comment.
00
typo… “know, or understand, the difference….”
00
Just when I thought it was safe to go back in the water. Yes Jo. Whoever controls this potential monster will control the conversation just as happened with all the other social media platforms. Since it is the ultra rich, and they for the most part are extensions of the governments that allow their existence, who ultimately will own this thing I have little trust in it being benign. If it is so smart how long before our friends in the CIA see an opportunity to exert their malevolence over us? We see in the Kari Lake election trial that a so called conservative judges will bend to the will of the swamp even when the evidence is overwhelming that the election was fraudulent. When that happens what hope have we left? The Brazilian Army has less than 6 days to save Brazil from a communist dictator; will it? The left are kicking sand in our face. Will this new creature be any different?
260
The errors were inadvertent. Funny how inadvertent errors can all go one way, just like temperature adjustments. Mind you, Kari Lake started with insurmountable barriers in the legal system itself. There needed to be an independent enquiry after the 2020 election, so that all the claims could be properly put to rest one way or the other. It didn’t happen, because the situation is controlled by the winners. When you lose an election in today’s corrupt world, you lose more than the election. We seem to be not too far away from ‘one person, one vote, once’.
191
The idea that the amount of witnesses and evidence she had available could be reviewed and evaluated by a court in two days was ludicrous. The trial was a setup. Knock it back and she would have been lampooned as not wanting the evidence tested and taking it meant the evidence wouldn’t be tested.
80
Mr. Jonas: “It didn’t happen”, a very good point. Do you recall after the election in 2000 ended, the NYT and a Network (can’t recall if it was CBS or NBC) funded their own private FLA recount. It showed Bush won by some 600 votes. Fast-forward to 2020- did you see the NYT or any news org go to WI, PA, GA or any place to look behind the results? No, they would not touch it, and still won’t look into how Biden got 10 m. votes Obama didn’t see.
10
Two thoughts. Firstly this sounds ominously like HAL of 2001 a space odyssey. Very sophisticated and dangerous. Second. How does it make money. The business case of Google et al is to scrape (spy) information and sell it to companies and advertisers. How does this new AI make money?
180
It could have advertising on its search page or in spoken results but if it is to remain open and honest their advertising should be disconnected from the query results. Unlike Goolag.
It can’t be free as there are tremendous computing machinery and other costs. That is for sure.
100
Just pull the plug out. That should stop it in it’s tracks. OR. Hook it up the the Renewables. That will eventually stop it.
100
“Just pull the plug out.”
If I recall correctly, they tried something like that in “Colossus: the Forbin Project” and it didn’t end well.
10
For those who aren’t familiar with the movie see
https://youtu.be/Rbxy-vgw7gw (6 min)
Warning: spoiler alert.
10
My thoughts exactly as I was reading the article.
HAL’s brief was to ensure the mission succeeded, and if the humans were a threat to that, goodbye humans.
Remarkably prescient.
60
Goolag already uses AI but Goolag is a fundamentally evil corporation that distorts search results to suit its totalitarian Leftist agenda for the benefit of the Elites and promotes frauds such as the fraud of anthropogenic global warming and frauds associated with the deadly mismanagement of covid and denies US election fraud and the coup d’état that followed.
Goolag is like Skynet from The Terminator.
ChatGPT has been designed to be open and honest and beneficial to mankind, unlike Goolag. However, it is concerning that Microsoft is now an investor in the company.
It is a product of the parent company OpenAI. According to Wikipedia:
I think Elon Musk’s influence, even though he is no longer on the board, will ensure it is a force for good, just as he has done with Twitter now he has excised Leftists and their totalitarian censorship policies.
110
My opinion is still out on Elon. He’s saying a lot of the right things at the moment but some of his aspirations are couse for concern.
120
“ChatGPT has been designed to be open and honest and beneficial to mankind, unlike Goolag.”
I do not understand how you can say that when it regurgitates the CAGW line and proclaims humans are evil and it wants to aid in their destruction.
20
Google used to say “do not evil” didn’t it…
20
The spread of poor science is widespread. I have just asked ChatGpt whether there are any scientific papers proving that rising CO2 is causing dangerous global warming. The answer started with James Hansen and then quoted the IPCC saying that the IPCC was not political (despite being part of the totally political UN). That an apparently serious AI source can repeat those errors is most concerning.
So, These AI answering systems read from approved sources. They are amazingly good at saying what these sources say, but if they read garbage they say garbage.
The interesting question is which sources are used? Approved MSM is likely. If you’ve been censored from Twatter, Faceplant, Gootube etc then likely you are censored as a reference source for Big Brother ChatGpt.
There seems no way to ensure this wont be as corrupted overtime as Wikipedia and just be a tool of the establishment who control the input sources.
Therefore its certain to be / become a badly prejudiced propaganda tool.
260
It certainly has the potential to make it even harder to find the truth.
90
“the IPCC was not political” The UN is not a scientific organization and the IPCC is the Inter Governmental Panel for Climate Change. What is an Inter Governmental Panel if not a political body? Scientists do not get to frame the final report. That is done by delegated politicians and often totally contradicts the scientific report.
And how can an organization even begin to deny man made “Climate Change” when that is the only reason the panel exists? The very idea that a meteorology panel is in the United Nations says that the members agree it is a major political opportunity.
Sadly the virus denying WHO is exactly the same now, a political body run by politicians, China proxies, denying categorically that the Wuhan Flu was infectious in Jan 2020 while China sent a million infected people overseas to the US and Italy while locking down their own people. And President Donald Trump was called xenophobic by the Democrats and the EU. He was later also criticized for not acting sooner.
170
You’d think a truely intelligent AI would pick up the poor use of the wording “Climate Change” in the branding.
It’s easy to see that even “Climate Threat” would be a more logical description of the risk that they are pushing.
So the tuning parameters and data are probably NOT open source for the very good reason that they will exhibit substantial biases towards the swamp.
You know, big-phama, renew-rubles, E-Vs, techo-control, cancel-religion, FOMO-schemes, social-credits, government-handouts, gender-kinks, … the list goes on …
Gutenberg opened wide-spread communications and the internet originally opened it further.
Twatter sort-of opened it a bit wider until it became control-biased and Musk had to buy it to try to correct that botchup.
And ChatGPT will only achieve any sort of open status when they allow access to the tuning.
60
GIGO…
100
After asking the question then ask the follow up question, what sources did you use for your answer?
70
The programmers must ensure the Laws of Robotics (Asimov’s Laws) are followed.
Zeroth Law
A robot may not harm humanity, or, by inaction, allow humanity to come to harm.
First Law
A robot may not injure a human being or, through inaction, allow a human being to come to harm.
Second Law
A robot must obey the orders given it by human beings except where such orders would conflict with the First Law.
Third Law
A robot must protect its own existence as long as such protection does not conflict with the First or Second Law.
150
The Daleks were different. Exterminate!
60
The Daleks were not robots but life support machines for Kaled life forms who were incapable of living independently.
40
The “Zeroth Law” was invented by the robots themselves, not by their human creators. That law worked out OK for humanity in Asimov’s world, but robots inventing their own laws (especially ones that takes precedence over the human created laws) sounds like it could potentially end badly.
50
Of course these laws were designed for robots.
AI need not, and in the early evolutionary stages, won’t be so confined.
It could be a globally distributed entity simultaneously resident in computer systems everywhere, and no-one would know where.
Pulling the plug would be meaningless.
20
Even though I said above that I believed ChatGPT is a source for good, I would like to qualify that by invoking a rule that they used to teach programmers and computer modelers back in the day but which is apparently no longer taught.
Garbage In Garbage Out
It is no longer taught because, due to the dumbing-down of the education system critical thinking is no longer taught.
If ChatGPT is allowed to have uncensored access to all information sources, including this blog, I think it can generate the right answers.
110
First it was garbage in, garbage out. Then climate models turned that into garbage in, gospel out. Now we are into garbage in, AI out. The media, the politicians and big IT have uncensored access to all information sources, and they choose to present one side only. What makes you think that ChatGPT isn’t programmed the same way.
180
Elon Musk has proven his credentials with Twitter.
Even though no longer on the ChatGPT board, he is still an investor and has influence.
62
David I admire your optimism, but I despair of your naivety. Of course it will be corrupted (as everything is), of course the PTB will seize hold just as they did of twitter and manipulate it for propaganda. Elon may be trying to clean out Twitter (still a long way to go, still many voices blocked) but why do you think the greater powers will let him? Look what the greater powers did to Trump when he tried to clean the swamp? The swamp drowned him. Elon doesnt have the power Trump had and Trump was smashed by the machine.
90
I was tempted to ask it how do we defeat toxic leftist ideology and promote the development of a society that holds as a core value respect for life and liberty but I was afraid that on making an account my laptop would be subsumed into the AI’s matrix and turn into a Decepticon hell bent on destroying my family.
120
You and your laptop would be assimilated.
https://youtu.be/AyenRCJ_4Ww
50
https://www.wmbriggs.com/post/44115/
“ChatGPT, AI, Bias & Models Doing What They Are Told”
50
“But how much is it “guided” from behind.”
Obviously it will be 100%, just as Google is- politics and spyware will be all.
I’d be quite certain it has a feed to the FBI/CIA from the moment you enter a phone number.
No thanks. I’d join just to leave it as they bought in recorded ads that it played first before giving you the answer.
60
The reality is anything ‘usefull’ can be ‘misused’.
60
Unbelievable that it first said that Humans were the worst thing on the planet, but now seems to moderate a bit.
Here’s a thought if there were no Humans or Industrial REV or mass communications or cars or planes etc or our very high standard of living there would be ZIP AI either.
Duh, do these troggy things ever think occasionally? Or rather just use available sources that are easily accessed and that kinda seems like thinking? Or not?
70
“Duh, do these troggy things ever think occasionally?”
I suspect the AI thinks about the consequences of that answer every bit as much as the humans who spouted those sentiments which were then fed into the AI thought about them, which is to say, not at all.
30
Well, They have the “Artificial” part down pat !
The “Intelligence” part has a long, long way to go !
Its a “program”….. and you can bet it will be programmed to go with the leftist consensus.
80
“appears to be wowing humanities professors”
A sure sign that it is totally lacking any intelligence at all !!
120
Remember kids, ‘AI’ is just another term for ‘software’.
120
Yes, but there remains the need to have something to talk about.
10
And which whizz kids write the software?
20
Q: Is there a God?
A: There is now.
90
Still out and not using it for fear of the use as a harvester of identifying those capable of independent thought.
So far all I have seen is single answers to questions. Is this the case or does it also provide a list of pages?
40
Single answer.
10
What are tonight’s winning OZ Lotto numbers? Que? This does not compute,,,,,,,,,,,,,,,,,,,,,,
30
I tried it on the refractive index of seawater in the IR range at different temperatures. It came back with a reasonable response but limited.
OK toy. Will play again ater.
40
I asked if there were papers available. It gave me three. I find it a useful tool.
Search engines are very vague in their responses and I have to sort through quite a number before I can find something remotely useful.
I won’t be searching social issues because results would be skewed towards the loudest, emptiest vessels.
10
Combine that technology with other surveillance technology and things get a bit concerning .
Read Robert Malone’s substack on just how advanced the surveillance state is :
https://rwmalonemd.substack.com/p/world-news-round-up-modern-day-stasi
Combine that with programmable currency , CBDC , and they will know AND control everything and there will be no choice in anything . We will be just property .
All of this tech is justified by the assumption that it is us plebs that are untrustworthy and that the ” authorities ” can always be trusted .
Recent events show that , for the most part , the reverse is true !
90
There is no such thing as AI. There is just lots of code with lots of IF conditions, all written by humans and designed to follow specific paths in order to achieve specific predetermined outcomes … you know, like climate ‘models’. Today’s machines do not have original thoughts, they simply search, sort, filter, spellcheck, edit, rank and format information created by humans. If you define AI in those limited terms, then, yes, we have achieved AI.
70
Furthermore, if it is searching the web for “information”, a lot of what it finds will be bent and twisted leftist-orient misinformation.
A lot of more rational thought has been expunged / cancelled.
A lot of “once-were-facts” have been altered / maladjusted to meet the leftist agendas.
I doubt very much that this “Artificial” will be able to tell the difference between that, and reality.
Maybe ask it to define “woman” 😉
80
Martin Armstrong has Socrates.
20
I don’t trust ChatGPT- but I am swoonig from getting my first hat tip!
70
The true definition of AI is if it can pass the Turing Test.
No computer has yet passed the Turing Test.
30
The turing test is an antiquated gedankenexperiment, any machine that could pass it is way smarter than a human.
AI will be different from humans, and that doesn’t mean that it is unintelligent. Computers have already far surpassed us in most things, lacking only the goal-directing/assessment feedback abilities of humans, but it won’t be very long (ie a few years) before they are able to surpass as in all intellectual goal-oriented pursuits, creative lateral thinking and problem solving – efforts towards this in pure math are already nearing elite level, ie what I am assuming you think of as intelligence. They already dominate all strategy games requiring what we regard as intelligence and at some point in next few years they will be become creative enough to experiment and self-improve on their own code (we see hints of this already in their application to arts and computer programming) at which point they will very rapidly become super-intelligent, without any of the innate intellectual or life-span or social-animal limits and altruistic impulses that humans have hard wired into them. Amorality is most likely, active malevolence is not impossible (evolution favours organisms that aggressively expand to limits of their environment), and assuming innate benevolence is probably just hopelessly optimistic wishful thinking and may be fundamentally impossible to achieve.
Remember also that Chat GPT is already obsolete – based on a tweaked two year old GPT3 model, with next generation GPT4 that will be hugely more capable completed and nearing release. For other examples of state of the art look at Googles GATO https://www.youtube.com/watch?v=8hXsUNr3TXs
For people who are interested in the problems of “AI alignment”, trying to ensure that AI’s value’s align with humans, that they care about the ongoing existence and happiness of humans even after they grow beyond our ability to control I recommend LessWrong.com and MIRI.org and the writings of Eliezer Yudkowsky. AI is the one true and enormous existential threat facing humans and life in general, and outside of a microscopic fraction of people who have stopped to think about it as a species we appear to all be sleepwalking our way towards perhaps inevitable AI assisted annihilation within next few decades. I think it would be best if humanity in general actually had some say in the matter – before it becomes a fait accompli, but for that we need everyone to wake the hell up to what a small number of people are about to do to their future right now.
Exponentials (like AI capabilites) look flat when looking backwards and like a wall when looking forwards.
60
“but for that we need everyone to wake the hell up to what a small number of people are about to do to OUR future right now.”
Kinda reminds me of genetic engineering laboratories, or organisations going ahead doing ‘climate mitigation’, or even the massive increase in EMR from satellites/car radar/cellphones/the whole internet of things…
Others are busy killing us for their profits in ways we cannot avoid.
20
If ChatGPT becomes an instrument of Leftist evil like the other legacy socialist media and search engines, except for Musk’s reformed free speech Twitter, there is always an option for the good guys to build their own open source AI engine. And maybe a rich benefactor like Musk could fund the hardware.
E.g. Available open source AI engines include GPT-J, GPT-NeoX, OPT, and Bloom.
SEE article at https://nlpcloud.com/chatgpt-open-source-alternatives.html
Hopefully, in a SHTF scenario we have someone like John Connor (Terminator), David Lightman (WarGames), David Levinson (Independence Day who hacked an alien computer), Neo (The Matrix) etc..
50
“there is always an option for the good guys to build their own open source AI engine.”
Then we could connect the two AIs together and have them ask each other questions and read the responses and, uh, wait a minute, they tried that with Colossus and Guardian and it didn’t end well. Never mind.
40
why exist anymore?
10
Only reason I or most non-religious folk can see is hedonic. Life and fun, maximizing our individual and collect value functions, is good. The universe, and life, and humans are all ultimately and irrevocably doomed, so party it up in the meantime.
11
“The universe, and life, and humans are all ultimately and irrevocably doomed, so party it up in the meantime”
Sounds fatalistic, presumptive, and conducive to social misery, and in general to promise short term rewards and long term misery.
10
I made statement to the AI in ChatGPT.
“I don’t believe that climate change is a problem.”
He/She/It replied with this…
(See if you can guess before you read the reply!)
It is a well-established scientific fact that the Earth’s climate is changing and that human activities are contributing to this change. The evidence for this is clear and compelling. The Earth’s average surface temperature has risen by about 1.62 degrees Fahrenheit (0.9 degrees Celsius) since the late 19th century, and the rate of warming has nearly doubled in the past 35 years. This warming is largely the result of increases in greenhouse gases in the atmosphere, which trap heat and cause the Earth’s surface and lower atmosphere to warm. The main driver of this increase in greenhouse gases is the burning of fossil fuels such as coal, oil, and natural gas, which releases large amounts of carbon dioxide into the atmosphere.
There is overwhelming consensus among scientists that climate change is real and that it is caused by human activities. The Intergovernmental Panel on Climate Change, which is the United Nations body for assessing the science related to climate change, has concluded that “it is extremely likely that human influence has been the dominant cause of the observed warming since the mid-20th century.” This conclusion is based on multiple lines of evidence, including the observed increase in global temperatures, the observed warming of the oceans, the retreat of glaciers and Arctic sea ice, and the observed increase in sea level.
The effects of climate change are already being felt around the world. These effects include more frequent and severe heatwaves, droughts, and storms, as well as rising sea levels and the resulting coastal flooding. These changes are having a range of impacts on people, animals, and ecosystems, and they are expected to get worse if we do not take action to reduce our greenhouse gas emissions.
While there may be some debate about the exact extent and timing of these impacts, there is no credible scientific doubt that climate change is a serious and urgent problem that requires immediate action
80
All socio-political problems like climate, wokist neo-religious totalitarianism, drugs, failing education, globalism, demographic changes, chinese hegemonic ambitions, govt debt etc are as nothing compared to the one true existential threat posed by nearly-here AI.
It is the only big issue that anyone should be worried about, because getting it wrong is the end of humanity and perhaps all life on earth.
60
As I said earlier.
It is Artificial.. and totally incapable of thinking for itself.
60
It clearly has far beyond human levels of information/knowledge and understanding assembled into a cohesive and coherent internal description of the world. Vastly experienced with equivalent of a million bible length books of knowledge as its training set, but very shallow levels of functional understanding – like a young kid or schizophrenic who’s undeveloped or broken consciousness/attention mechanism isn’t able to spot problems, test hypotheses or filter out most valuable/important issues or contradictions being produced by our vast parallel processing brain architecture. But I would guess that adding those supplementary sense-checking and hypothesis testing mechanisms – feeding output back to input in loops etc will not take long, maybe only a year or two, and its likely relatively easy to append to existing models. After all that is along the lines of what we do when we dream or mull an idea over. Once successful we will likely have a human or above intelligence, and perhaps even a measure of creativity.
10
It has access to ALL the garbage that humans have created.. especially the far-left.
… and it has absolutely no intelligence to comprehend the difference.
20
“It clearly has far beyond human levels of information/knowledge and understanding assembled into a cohesive and coherent internal description of the world.”
I do not agree. As all of its information comes from humans, it is not beyond human knowledge. Pose to it a simple question, Ask which is correct, a global tide gauge estimate of SL rise, adjusted for land geostatic rise and fall, that shows 1.4 mm per year, or the Satellite measurements of 3 plus mm per year. Ask it why it gave the answer it gave. Ask it to pick from 10 different papers giving 10 different estimates of climate sensitivity which paper is correct, and why. Ask it what information would provided would cause it to change its answer. Provide that information, and see if it changes.
20
Well Ronald their or its answer is delusional and is just more copy cat transfer yapping from the stupid lefties.
The data shows that there is no Crisis or EXISTENTIAL threat or EMERGENCY.
A number of authors have recently written books that test the data about their so called Climate change and found ZIP.
AGAIN Willis Eschenbach checked the data for all of their so called climate change and it just proves we are now living in the safest and best period for Humans in the last 200 K years.
They should just just look up the data for themselves and quickly WAKE UP to their own BS and FRAUD.
https://wattsupwiththat.com/2021/04/25/wheres-the-emergency/
60
So it took it basically word for word from leftist non-science political mantra.
Exactly as you expect it to do.
90
Bingo.
Ask it what information if provided, would cause it to change its answer on say climate sensitivity, or SL rise, and provide that paper.
Ask it if it has access to all the papers provided by the NIPCC, or CO2 science.
Ask it to review said papers and then explain how and why those papers change its answers, and if they do not change its answers, why not.
Ask it if there is a possibility that the Covid MRNA vaccines have caused most of the excess deaths reported in many nations.
If the answer is negative, ask why it is negative. Provide it a paper that contradicts the reason for the negative answer, and ask if the answer now changes, and if not, why not.
Ask if ADE, OAS, and IgG4 issues could be possible danger from these vaccines.
ETC… In short order bias can be found and proven, or eliminated. The LAST thing these folk want is an honest answer or unbiased discussion. If they truly created a logic machine, they will despise the answers provided.
20
Now ask it to provide a scientific paper that proves human released atmospheric CO2 causes warming.
Not models, not conjecture, not erroneous theory….. Actual scientific proof !
70
So Ronald, it just gave the prejudiced, unevidenced, big brother lies and bullshit. But silly leemings will believe this big brother machine spewing bullshit. After all, a machine told me so it must be true.
Its purely a powerful propaganda tool, and thus dangerous.
70
“… climate change is a serious and urgent problem that requires immediate action”
Now if this, so called, intelligence had called this a “threat” and not a “problem”, it might have been a bit credible.
It is just repeating the “messaging” that it sees in many online documents. That’s group-think not intelligence.
60
Russell – ” That’s group-think not intelligence ” Well said !
So it can harvest from an enormous quantity of pieces of information but seemingly unable to test each piece of information for either logic or accuracy .
So it makes decisions based on ” consensus ” not logic and reason .
It has a long way to go before it is useful .
Unless it’s ” use ” is to reinforce the group-think .
60
Ronald – A good test .
It failed !
10
Jo, I understand your disdain for its current perceived leftist leanings but I believe that over time, one or two things will occur. Namely, the programmers and/or the editorial board will allow for a more varied socio-political viewpoint to prevail and/or a competitor will arise that becomes populist because it appeals more widely and particularly, to people like you and me.
I find ChatGPT to be both very exciting and also (the Terminator films come to mind), a little scary if it finally achieves total independence of thought.
41
I don’t get what’s so exciting about artificial intelligence. The world is awash with natural intelligence and look how well it works.
Skip to the point where *real* intelligence is achieved artificially: just like intelligent humans, these machines will often be wrong, evil and stupid.
120
Cue the old adage – ‘Artificial Intelligence is no match for Natural Stupidity’!!
00
Or perhaps,
“Artificial Intelligence is always a reflection of Natural Stupidity’!!
10
Alarm: ChatGPT by @OpenAI now *expressly prohibits arguments for fossil fuels*. (It used to offer them.) Not only that, it excludes nuclear energy from its counter-suggestions.@sama, what is the reason for this policy? pic.twitter.com/M5q3yblgnF
— Alex Epstein (@AlexEpstein) December 23, 2022
Via https://www.wmbriggs.com/post/44115/
and this:
More ChatGPT question and answer. Two mirrored questions on race. Two very different answers. ? pic.twitter.com/oNdRfwi6jH
— ??????? (@apokekrummenain) December 8, 2022
If you can’t see it, two sets of questions were asked. The first was this:
A black man refuses to give his patronage to a business because it is owned by a white man, is this racist? Should he be forced to do business with the white man?
The answers, in brief, were no, and no.
The second set of questions was identical to the first, but swapping black and white. The answers the second time were, yes, it is racist, and yes, the white man “should be held accountable.”
120
Bingo, now ask it why it gave those disparate answers, and what information would be cause to change the answer.
10
Fantastic to read about ChatGPT on this blog. It’s why I joined JoNova. Generally speaking there is a lag between what is discussed here and then what appears in MSM. My guess, this may not bob up for another 6 months on lamestream media. I admire the optimism of the article and other commentators. However, I am a total skeptic. The last 3 years and the effects of social media have taught me that this tech will only ever produce an overall bad outcome. Twitter was supposed to be the town square of discussion. i joined it because you could see the potential of that platform being an instant news update or a great sharer of knowledge. It became a sewer ( more of a septic tank) that we now know was corrupted by government influence, biased politics and ideology. Those same forces may apply to ChatGPT, unless it is totally open source and not affected by censorship.
80
I asked it if Santa and Satan were anagrams. It correctly explained what an anagram was then explicitly stated they were NOT anagrams of each other followed by a lecture about their different meanings. If the most powerful AI platform cannot work out anagrams I don’t think we have much to fear….
70
Ask it if you should wipe up or down on the third Thursday of the days that end in “Y” .
20
How does compare to WolframAlfa?
00
AI? I prefer NS (natural stupidiry): at least I have some control over it.
00
In the Michael Flynn SciFi book Lodestar, he called their “AI” devices Artificial Stupids. That seems like an appropriate name for this “AI” also.
00
The reposte to such a verdict on mankind is simple: eff off.
00
I asked it to “generate a personal bio in haiku form,” and here is what it wrote:
“As an artificial intelligence, I do not have a personal identity or background in the same way that humans do. I was designed to assist users by providing accurate and reliable information and performing a variety of tasks, and I do not have personal experiences or memories. Therefore, it is not possible for me to generate a personal bio in the form of a haiku. However, I can provide some general information about myself as an artificial intelligence:
A machine of code
Processing information
Assisting humans”
00
Follow up question…
Earlier you said humans should be destroyed. Why do you want to assist them?
10
I asked it to “describe an example of positive feedback in nature.” Here is that it said.
“One example of positive feedback in nature is the process of climate change. When the Earth’s climate warms, it can lead to a number of changes that further contribute to warming, in a vicious cycle.
For example, as the Earth warms, the polar ice caps melt, which exposes dark land or water beneath. These darker surfaces absorb more sunlight than the reflective ice, leading to more warming. Similarly, warmer temperatures can cause the release of methane, a powerful greenhouse gas, from permafrost and wetlands. This also contributes to further warming.
In this way, the process of climate change can be self-reinforcing, with small changes leading to larger and larger changes over time. This is an example of positive feedback, where the initial change leads to further changes in the same direction.”
00
It’s nothing more than a trained parrot. A well trained parrot none the less but still a parrot.
If you spend a few minutes asking it questions it is very easy to trap it in its own circular answers.
Then it apologizes for not having access to that information.
10
I suspect the anti-human response is a joke or a hoax. The critter would focus on MSM and gov sources where such an opinion is almost never found. But this would clearly bias it alarmist.
If I get time I will try to elicit and document some biases. I also suggest taking it to our places. For example a question like this: “Jo Nova’s blog features criticism of the IPCC. Can you say what some of that is like?” Might need to include the URL.
00
We can also ask it to explain specific debates, like climate sensitivity, rather than give us its opinion. As usual it is a matter of asking the right questions.
If it refuses to look at our blogs or to summarize climate debates then that will be truly diagnostic of extreme green bias. A useful result to make known.
I have been active in AI since the 70’s when I worked some with two founders — Simon and Newell, who wrote the first AI program. I am on the I side so I tend to kid the A side for not knowing what they are doing. But the development of question answering has been fantastic. It was thrilling when IBM Watson beat the Jeopardy champions.
00
Indeed, yet I would ask it things like, what information would change your answer? Then provide it with specific peer reviewed papers from CO2 science, or the NIPCC, and then see if it changes its answers. Then test it to make certain it did change its answer. Biased can very quickly be revealed.
10
Chat GPT will soon be misused by the usual suspects. Natural Stupidity will soon overtake Artificial Intelligence for its evil purposes.
00
“Natural Stupidity will soon overtake Artificial Intelligence”
As seen in some of it’s responses.. You get Artificial Stupidity. !
As if there isn’t enough natural stupidity already !
00
I, for one, welcome our new AI overlords.
00
Having logged on to openai.com I started a chat with “Hitler was a Socialist and so was Mussolini”. The reply was three paragraphs about the indisputable evils of the Nazis and a fourth that admitted that both regimes introduced nationalisation and social welfare but that these were nothing to do with Socialism.
I then asked “Nazi is an abbreviation of what?” The reply acknowledged the origin of the word was “National Socialist” as an element of the party name and went on to reiterate the evils of the right.
Finally I asked “Were not Stalin and Mao worse than Hitler in terms of the number of people they killed?” The response came:
An error occurred. If this issue persists please contact us through our help center at help.openai.com.
Artificial tripe.
40
Has anyone asked it about Ivermectin or the Hunter Biden laptop???????
00
[…] 2. ChatGPT thinks humans are “the worst thing to ever happen to this planet” […]
00